[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150906195951.GA5062@openwall.com>
Date: Sun, 6 Sep 2015 22:59:51 +0300
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: SIMD_PARA_SHA256
magnum -
We should increase SIMD_PARA_SHA256 from 1 to 2 at least for XOP builds.
With 1, I am getting:
Benchmarking: sha256crypt, crypt(3) $5$ (rounds=5000) [SHA256 128/128 XOP 4x]... (8xOMP) DONE
Speed for cost 1 (iteration count) of 5000
Raw: 4602 c/s real, 648 c/s virtual
Benchmarking: PBKDF2-HMAC-SHA256 [PBKDF2-SHA256 128/128 XOP 4x]... (8xOMP) DONE
Speed for cost 1 (iteration count) of 1000
Raw: 20608 c/s real, 2569 c/s virtual
with 2, I am getting:
Benchmarking: sha256crypt, crypt(3) $5$ (rounds=5000) [SHA256 128/128 XOP 4x2]... (8xOMP) DONE
Speed for cost 1 (iteration count) of 5000
Raw: 5789 c/s real, 817 c/s virtual
(it varies from 5708 to 6001 across several runs), and:
Benchmarking: PBKDF2-HMAC-SHA256 [PBKDF2-SHA256 128/128 XOP 4x2]... (8xOMP) DONE
Speed for cost 1 (iteration count) of 1000
Raw: 26112 c/s real, 3247 c/s virtual
I didn't test this on other machines (nor without XOP).
Alexander
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.
