[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 04 Aug 2012 23:42:05 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: sunmd5
On 2012-08-04 23:21, jfoug wrote:
>> From: magnum [mailto:john.magnum@...hmail.com]
>>
>>> Also, each round, there is copying from a flat buffer, back into
>>> a MMX_COEF buffer, for each candidate. I would love to avoid
>>> this, but since you have no idea if a candidate will use the 1
>>> block, or the 25 block crypt on any given round, I do not see any
>>> way around that issue.
>>
>> That sounds expensive. Some cleverness here might be the key.
>
> I agree, but I see no way.
>
> You have:
>
> Outs[1024]
>
> In[PARA*COEF]
> InBig[25][PARA*COEF].
Could you not just use one In[25][PARA*COEF] and always use that? I.e.
for 1-limb runs you just use In[0]. Maybe that would not help much
though. And on another thought I guess that would mean we blow the first
part of the long constant. On the other hand we only need to repair it
once per crypt_all(). Oh well, I guess I should look at the code, or at
least the format, before coming up with confusing ideas :-P
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.
