[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ae6dfe4586efe049d1e567ec6ecd4f76@smtp.hushmail.com>
Date: Thu, 29 Aug 2013 23:51:12 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Sayantan Weekly Report #11
On 27 Aug, 2013, at 7:34 , Sayantan Datta <std2048@...il.com> wrote:
> On Tue, Aug 27, 2013 at 2:37 AM, magnum <john.magnum@...hmail.com> wrote:
> 57M for SHA1 is a bit disappointing - running wpa-psk I get 106M (sha1/s).
>
> Speed is okay(almost same as before) on my 7970 even after the patch. Maybe the patch is affecting weaker GPUs only.
I meant it's poor on this GPU regardless of that bugfix patch. 57M is just three times faster than one CPU core.
Here's a theory: Since mask mode currently picks 26*26 == 676 as multiplier in this case, I have no choice other than to lower GWS all the way to 16384 (ending up in a KPC of 11075584) to keep kernel duration short enough to avoid triggering watchdogs and rebooting the whole thing. I presume the "low" GWS is the reason for the poor performance. Maybe it would be way better with a multipler of 26 and a GWS of 524288... in *this* case.
Maybe the kernel can be improved (reducing need for high GWS) but regardless of that, this means auto-tuning will be trickier: You need to figure out not only the GWS but the best *combination* of GWS and GPU-side mask that completes in less than 200ms.
magnum
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.
