[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Date: Mon, 27 May 2013 12:45:48 -0400
From: <jfoug@....net>
To: john-dev@...ts.openwall.com
Subject: Re: Initial work on cryptsha256 - SSE
I have checked this into bleeding, along with changes to sse-intrinsics.
It works pretty good in non-OMP mode. It is failing in OMP, and I will have to track down why.
Jim.
---- jfoug@....net wrote:
>
> ---- jfoug@....net wrote:
> > Not getting great improvement. Some, but not as much as I know I can get.
>
> Built on bull, and things look a little better.
>
> oSSL speed 350
> SSE2 speed 1080
>
> SSE4 fails, but I will likely be switching from flat mode, to mixed mode.
>
> Just fixed GATHER, and I get 1580/s on SSE4 build on bull, with the existing 'flat' input buffers. Not too bad. The old GATHER was not working (only used right now in raw-sha256-ng-i if MMX_LOAD is not defined). I fixed GATHER, and created GATHER_x2 for the double width buffers I am using in sha256crypt.
>
> Jim.
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.
