[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20120623205222.GA1203@openwall.com>
Date: Sun, 24 Jun 2012 00:52:22 +0400
From: Solar Designer <solar@...nwall.com>
To: john-dev@...ts.openwall.com
Subject: Re: WPA-PSK fixes, OpenMP support
Lukas -
On Sat, Jun 23, 2012 at 10:39:00PM +0200, Lukas Odzioba wrote:
> Thank you for working on that. I tried to add omp support but my code
> introduced some bug and I sent magnum code without that.
Chances are that it exposed the bugs you already had. ;-) This is how I
found the two out of bounds writes that I mentioned.
> > BTW, where does the length 15 limit come from? Can/should we avoid it?
> It is something that I've got in mind and forgot to change. We can
> easily avoid that, even for gpu patches. I'll send fix for that.
Great.
> > Hmm, somehow "Optimal Group work Size" is different here.
> >
> > BTW, "group work size" sounds weird. Do we actually mean "global work
> > size" or "work-group size"?
>
> It should be work group size.
Can you submit a patch for that, for all formats that use the wrong
wording (not only yours)?
> Even better results we could get by overlapping cpu and gpu code ( we
> would utilize cpu AND gpu at one time), and there is a chance ( cpu
> time < gpu time )that sse-intrinsics.c won't be needed, (speedup
> won't be noticable because of low cpu usage) - but this thesis need to
> be verified.
Sure. My comments about sse-intrinsics.c was for the CPU-only format.
For overlapping CPU and GPU portions of code in WPA-PSK and MSCash2
formats, my planned support for multiple key blocks should help (that's
a formats interface enhancement).
Thanks,
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.
