[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 11 Apr 2023 09:45:15 +0200
From: magnum <magnumripper@...hmail.com>
To: john-users@...ts.openwall.com
Subject: Re: John the Ripper efficiency
On 2023-04-07 17:29, Claudio André wrote:
> Will run 2 OpenMP threads
> Benchmarking: sha512crypt, crypt(3) $6$ (rounds=5000) [SHA512 128/128 SSE2
> 2x]... (2xOMP) DONE
> Speed for cost 1 (iteration count) of 5000
> Raw: 1203 c/s real, 602 c/s virtual
>
> Benchmarking: sha512crypt, crypt(3) $6$ (rounds=5000) [SHA512 128/128 AVX
> 2x]... (2xOMP) DONE
> Speed for cost 1 (iteration count) of 5000
> Raw: 1657 c/s real, 829 c/s virtual
>
> Benchmarking: sha512crypt, crypt(3) $6$ (rounds=5000) [SHA512 256/256 AVX2
> 4x]... (2xOMP) DONE
> Speed for cost 1 (iteration count) of 5000
> Raw: 3068 c/s real, 1541 c/s virtual
>
> Benchmarking: sha512crypt, crypt(3) $6$ (rounds=5000) [SHA512 512/512
> AVX512F 8x]... (2xOMP) DONE
> Speed for cost 1 (iteration count) of 5000
> Raw: 8092 c/s real, 4056 c/s virtual
I didn't look into this for a while. Ideally AVX2 should be 2x the speed
of AVX.
AVX512F includes the neat instruction vternarylogic() which is the same
as nvidia's LOP3.LUT - it speeds up SHA-2 (and others) considerably and
that's why it's more than double the speed of AVX2.
magnum
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
