Am 03.07.2015 um 03:33 schrieb Mobile Mouse:
> 1. My opinion - yes to BLAKE2 and Skein, not sure about McBits and
> PHC, abstain wrt. crypt and scrypt.
Sorry forgot two (three) on my list:

  * ChaCha (we already have salsa20, so shouldn't be too hard)
  * Poly1305
  * StreamCipher + Poly1305 (AEAD cipher, with stream cipher templatizable)

McBits isn't a must-have, although it looks a lot easier to implement
than standard Niederreiter or McEliece.

I'd do Skein (+ Threefish as underlying construction), but it's
inconvenient for me at the moment, I'll go through the my CryptoJPM
files and propose a merge request on this list for the Skein+Threefish
(Skein: full packet, hash, mac, KDF). ETA: September '15, (as soon as
there's a Skylake R13)
I'll get Visual Studio 2015 around the same time and will see if I can
fix the SSE related errors occurring with VS2012 (meaning I can provide
scrypt + BLAKE2 family of MAC & hash).
I think it'll also be possible to me to dive into the Salsa20 code and
adapt it for ChaCha.

I think as soon as PHC has winner(s) (Q3 '15?), we should implement
them, because they'll be (far) better than (even) scrypt and bcrypt and
our current only choice PBKDF2 (!).
>
> 2. Yes, it is Kalyna (Ukrainian vs Russian pronunciation :). Timing
> attacks could probably be mitigated - the logic of doing so is
> reasonably well-known... Also, I personally find cache-timing attacks
> rather exaggerated, for symmetric ciphers at least. Now, AES has a
> distinct advantage here, because Intel put it in hardware - good luck
> timing cache :). Kalyna or any other software-based implementation
> wouldn't be so lucky, so more work for us, and worse performance...
>
Again, I certainly won't implement it. If you can implement it
/securely/ I think we can include it as it's the libraries aim to
provide as many algorithms as possible - in a high quality manner.

BR

JPM

> Sent from my iPad
>
> On Jul 2, 2015, at 17:27, Jean-Pierre Münch <[email protected]
> <mailto:[email protected]>> wrote:
>
>>
>>
>> Am 02.07.2015 um 23:21 schrieb Mobile Mouse:
>>> On Jul 1, 2015, at 20:35 , Jeffrey Walton <[email protected]
>>> <mailto:[email protected]>> wrote:
>>>> > I oppose adding Cmake support. I have it in one of my projects at
>>>> work, and it’s nothing short of nightmare.
>>>>
>>>> OK, points taken.
>>>>
>>>> How do you feel about providing it at the Patch page
>>>> (http://www.cryptopp.com/wiki/Category:Patch)? This way, its
>>>> available to those who want it, but it falls into the "provided,
>>>> unknown quality, and maybe not maintained" category. (This assumes
>>>> we will actually get our hands on a CmakeList.txt :).
>>>
>>> My concern is that if it is there - a bonehead is bound to pop up
>>> that would create something useful and make it dependent on cmake.
>>>
>>> Repeating myself - there are enough of useful things that are
>>> missing to inundate this project with something of adverse (IMHO)
>>> usefulness.
>> A short list ?:
>>
>>   * Skein
>>   * BLAKE2
>>   * bcrypt
>>   * scrypt
>>   * McBits <https://eprint.iacr.org/2015/610.pdf> (?)
>>   * PHC-winner(s) (TBA)
>>
>>>
>>> For example, how about adding support for Kalina block
>>> cipher? https://eprint.iacr.org/2015/650.pdf
>> 1. It think it's "Kalyna".
>> 2. Who should implement this? I certainly couldn't. The problem with
>> this cipher is that it is AES-like meaning we (may) have to fear
>> cache-timing attacks, which have been mitigated for AES, but for
>> Kalyna? I think we could implement, but maybe we shouldn't because
>> chances are we're getting it wrong (only speaking for myself though).
>>
>> BR
>>
>> JPM
>>
>>> -- 
>>> -- 
>>> You received this message because you are subscribed to the
>>> "Crypto++ Users" Google Group.
>>> To unsubscribe, send an email to
>>> [email protected].
>>> More information about Crypto++ and this group is available at
>>> http://www.cryptopp.com.
>>> ---
>>> You received this message because you are subscribed to the Google
>>> Groups "Crypto++ Users" group.
>>> To unsubscribe from this group and stop receiving emails from it,
>>> send an email to [email protected]
>>> <mailto:[email protected]>.
>>> For more options, visit https://groups.google.com/d/optout.
>>
>> -- 
>> -- 
>> You received this message because you are subscribed to the "Crypto++
>> Users" Google Group.
>> To unsubscribe, send an email to
>> [email protected]
>> <mailto:[email protected]>.
>> More information about Crypto++ and this group is available at
>> http://www.cryptopp.com.
>> ---
>> You received this message because you are subscribed to the Google
>> Groups "Crypto++ Users" group.
>> To unsubscribe from this group and stop receiving emails from it,
>> send an email to [email protected]
>> <mailto:[email protected]>.
>> For more options, visit https://groups.google.com/d/optout.

-- 
-- 
You received this message because you are subscribed to the "Crypto++ Users" 
Google Group.
To unsubscribe, send an email to [email protected].
More information about Crypto++ and this group is available at 
http://www.cryptopp.com.
--- 
You received this message because you are subscribed to the Google Groups 
"Crypto++ Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply via email to