Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAKGDhHXA3Azr=UBdjqtV0gPhj=-Pxmgytjf4aPpUBn5kTR8cig@mail.gmail.com>
Date: Tue, 6 Oct 2015 00:42:53 +0200
From: Agnieszka Bielec <bielecagnieszka8@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Agnieszka's GSoC formats

2015-09-24 21:38 GMT+02:00 magnum <john.magnum@...hmail.com>:
> On 24/09/15 20:59, magnum wrote:
>>
>> On 24/09/15 19:59, Agnieszka Bielec wrote:
>>>
>>> btw. I repaired rest of the formats which I broked down. dynamic(md5)
>>> fails but also fails on my laptop witchout any of my changes.
>>
>>
>> Great, I'll try it out and see if I can nail that md5 problem.
>
>
> Argon2i segfaults on both super and well (fglrx CPU device):
>
> Device 3: Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz
> Testing: argon2d-opencl [Blake2 OpenCL]...
> memory per hash : 1.50 MB
> PASS
> Testing: argon2i-opencl [Blake2 OpenCL]...
> memory per hash : 1.50 MB
> Segmentation fault

I modified the code, --dev=3 doesn't segfaults but on super --dev=6
and my card is slower

on my card it's 1882 c/s -> 1777 c/s
on --dev=6 it's 6678 c/s -> 5907 c/s

it segfaults when I'm casting an int array to long array
but also when I'm creating array as long, next I'm casting it to int
and again I'm casting to long

so I don't know if commit this or not

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.