[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 10 Nov 2011 16:11:41 -0600
From: "jfoug" <jfoug@....net>
To: <john-dev@...ts.openwall.com>
Subject: RE: more targets using sse-intrinsics.S
>From: magnum [mailto:john.magnum@...hmail.com]
gets using sse-intrinsics.S
>
>Looks good. While the intrinsics file is a little slower than md5-mmx.S
>for SHA-1, it has the great advantage of being thread safe.
I think a 'good' build of sse2-i is very nice. It seems faster in most things than the hard coded 32 bit .S stuff.
As for SHA1, possibly it is the LARGE incrementally used temp buffer that is causing the slowdown. Since it is the compiler putting out code, possibly we could get the SHA1 intrinsic stuff running faster by using temp_var % 16 set of buffers (instead of the 'full' 80 buffers).
> Did you try any OMP builds on Windows?
I have not yet. I am still building, and testing things for all build rules, prior to trying a OMP set.
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.
