[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 28 Jun 2012 11:58:15 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Re: Re: Failed self test for raw-sha1-ng (linux-x86-sse2i
OMP)
On 2012-06-28 11:45, Tavis Ormandy wrote:
> magnum <john.magnum@...hmail.com> wrote:
>> It got a little slower benchmark now, hopefully just because of new test
>> vectors.
>>
>
> Let me see if I can win that back :-)
>
> Re-ordering the W loads closer to first use seems to help gcc allocate the
> registers a bit. I'm not 100% happy with the code it generates, but it looks
> a little better. I think I can also nudge up the number of crypts with the
> cmp_one changes.
>
> On my linux-x86-native (xeon) machine, before:
>
> $ ../run/john --test=30 --format=raw-sha1-ng
> Benchmarking: Raw SHA-1 (pwlen <= 15) [128/128 SSE4.1 intrinsics 4x]... DONE
> Raw: 12730K c/s real, 12803K c/s virtual
>
> after:
>
> $ ../run/john --test=30 --format=raw-sha1-ng
> Benchmarking: Raw SHA-1 (pwlen <= 15) [128/128 SSE4.1 intrinsics 4x]... DONE
> Raw: 13103K c/s real, 13160K c/s virtual
>
> Please verify you get similar improvement, because I havn't tried it on x64.
>
> Push request sent.
It got a tad better so I committed it.
Yesterday's version:
Benchmarking: Raw SHA-1 (pwlen <= 15) [128/128 SSE2 intrinsics 4x]... DONE
Raw: 13431K c/s real, 13431K c/s virtual
After your bug fixes and new test vectors:
Benchmarking: Raw SHA-1 (pwlen <= 15) [128/128 SSE2 intrinsics 4x]... DONE
Raw: 12695K c/s real, 12695K c/s virtual
Now:
Benchmarking: Raw SHA-1 (pwlen <= 15) [128/128 SSE2 intrinsics 4x]... DONE
Raw: 12802K c/s real, 12802K c/s virtual
And if I restore the test vectors from yesterday's version, its:
Benchmarking: Raw SHA-1 (pwlen <= 15) [128/128 SSE2 intrinsics 4x]... DONE
Raw: 13265K c/s real, 13265K c/s virtual
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.
