Openwall GNU/*/Linux - a small security-enhanced Linux distro for servers
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 24 Sep 2015 22:09:50 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Agnieszka's GSoC formats

On 24/09/15 21:55, Solar Designer wrote:
> On 24/09/15 21:38, magnum wrote:
>> Argon2i segfaults on both super and well (fglrx CPU device):
>
> BTW, by fglrx here, you probably mean APP-SDK, which is currently part
> of the fglrx distribution, but can also be downloaded and installed
> separately.  Also, the code in jumbo to detect Catalyst version appears
> to actually detect the APP-SDK version and derive the likely Catalyst
> version from that, so it may misdetect Catalyst version if APP-SDK was
> installed separately on a system that also has fglrx.
>
> Does this understanding sound correct to you?

Maybe. I call it fglrx because that's what I install as a package in 
Ubuntu (although normally downloaded from AMD). With just those packages 
in a fresh install, it's a mess already: They present various totally 
unrelated version numbers (none of which reflecting a "Catalyst version").

Catalyst 14.9 seem to consist of fglrx version 14.301 and "something 
else" version 1573.4 (iirc there are even more version numbers 
involved). It's probably all really obvious to that poor AMD guy but 
it's a damn mess for me as a user. People refer to "Catalyst x.y" and 
you just can't know what you have from looking at your version numbers, 
or often not even in the file names you installed from.

On a side note, running oclHashcat with APP-SDK installed is not even 
supported. Not sure how they run it themselves...

magnum

Powered by blists - more mailing lists

Your e-mail address:

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