[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CALW8-7JDeypFeizHf3VsBdrsTDkYitU5VCrg=nL6RZrgSizV6A@mail.gmail.com>
Date: Wed, 30 Sep 2015 15:06:21 +0200
From: Dmitry Khovratovich <khovratovich@...il.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: Re: [PHC] Specification of a modular crypt format
Thanks for comment!
I am now polishing the spec so feel free to throw in more.
On Wed, Sep 30, 2015 at 2:38 PM, Alexander Cherepanov <ch3root@...nwall.com>
wrote:
> On 2015-09-28 14:50, Dmitry Khovratovich wrote:
>
>> On Sun, Sep 27, 2015 at 11:20 PM, Alexander Cherepanov <
>> ch3root@...nwall.com> wrote:
>>
>>> Hm, there seems to be some indeterminism in Argon2 itself here? According
>>> to spec: "The actual number of blocks is $m'$, which is $m$ rounded down
>>> to
>>> the nearest multiple of $4p$."
>>>
>>
>> To ease the hashing process, Argon2 allocates the same amount of memory
>> for each thread. Therefore, when you request 9 or 10 KB of memory and 2
>> threads, 8 KBytes will be used in fact. The output values will be
>> different, of course, as the requested 'm' is hashed as well.
>>
>
> Great!
>
> And about parallelism. It seems it's now allowed to be up to 255 but one
> place in the spec is not updated: "Parallelism. Argon2 may use up to 64
> threads in parallel, although in our experiments 8 threads already exhaust
> the available bandwidth and computing power of the machine."
>
> --
> Alexander Cherepanov
>
--
Best regards,
Dmitry Khovratovich
Content of type "text/html" skipped
Powered by blists - more mailing lists