Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 22 Oct 2013 20:28:12 +0200
From: Katja Malvoni <kmalvoni@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: ZedBoard: bcrypt

Hello,

On Mon, Oct 14, 2013 at 7:09 PM, Katja Malvoni <kmalvoni@...il.com> wrote:

> I will continue working on implementing multiple bcrypt cores using 100
> MHz clock and I'll return to clock problems after that is done.
>

This happened to be harder than I expected. I do have implementation of two
bcrypt cores which works in simulation. However, on ZedBoard one core works
fine while the other one produces incorrect result. I haven't managed to
figure out where is the problem yet, especially because both cores use the
same logic. Unfortunately, I encountered clock problems again. PlanAhead
reports maximum implemented frequency to be 84 MHz. But changing it had no
influence on outcome - first core works on 100 MHz clock while incorrect
result from the other core is the same for different clock frequencies. I
tried 83 MHz clock but it had same performance as 100 MHz clock which is
the same problem as earlier.
I'm using 50 MHz clock at the moment and I'll try to get other core working
on that frequency. When I succeed in that, I'll experiment with higher
frequencies.

Katja

[ CONTENT OF TYPE text/html SKIPPED ]

Powered by blists - more mailing lists

Your e-mail address:

Powered by Openwall GNU/*/Linux - Powered by OpenVZ