Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150725170013.GB2685@openwall.com>
Date: Sat, 25 Jul 2015 19:00:13 +0200
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: PHC: yescrypt on GPU

On Sat, Jul 25, 2015 at 06:17:06PM +0200, Agnieszka Bielec wrote:
> today I was removing getting smaller and smaller parts of the code to
> track down the slow part
> and this is indeed in pwxform, and when I have
>             x0 += p0[0];
>             x0 ^= p1[0];
> [...]
>             x1 += p0[1];
>             x1 ^= p1[1];
> 
> commented out speed is the same with copying and without

This makes sense.  In fact, the compiler might be optimizing out the
copying when the copied data is not used at all.

> (but I have another version of pwxform using vectors now)

Of course, you should be using 128-bit vectors (assuming the current
pwxform settings).

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.