Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 19 Aug 2015 04:47:32 +0200
From: Agnieszka Bielec <>
Subject: Re: PHC: Argon2 on GPU

2015-08-19 4:18 GMT+02:00 Solar Designer <>:
> Agnieszka,
> Am I correct that all of your work on Argon2 so far is on revision 1.0,
> without BlaMka and without the indexing function enhancement?  (I hope
> you're keeping track of relevant discussions on the PHC list.)

I have argon from github from
branch master. should I work now on 'enhcance' branch?

in argon2i.h
#define VERSION_NUMBER 0x11
in agon 2d.h
#define VERSION_NUMBER 0x10

I know that there is no blamka, no idea if there is indexing function
enhancement or not
I'm keeping track of discussion

> In other words, where and when did you obtain the Argon2 code that you
> integrated and ported to OpenCL?  What revision numbers did it have on it?
> On Wed, Aug 19, 2015 at 04:10:18AM +0200, Agnieszka Bielec wrote:
>> 2015-08-19 2:46 GMT+02:00 Solar Designer <>:
>> > On Tue, Aug 18, 2015 at 01:35:47AM +0200, Agnieszka Bielec wrote:
>> >> a@...l:~/f/run$ ./john --test --format=argon2d
>> >> Will run 8 OpenMP threads
>> >> Benchmarking: argon2d [Blake2 AVX]... (8xOMP)
>> >> memory per hash : 1.50 MB
>> >> using different password for benchmarking
>> >> DONE
>> >> Speed for cost 1 (t) of 1, cost 2 (m) of 1536, cost 3 (l) of 1
>> >> Many salts:     7760 c/s real, 971 c/s virtual
>> >> Only one salt:  7808 c/s real, 976 c/s virtual
>> >
>> > I assume it's just a temporary glitch that "Many salts" appeared
>> > slightly slower than "Only one salt" here.  They should be almost the
>> > same, or "Many salts" very slightly better.  In fact, once you're done
>> > debugging these formats, you'll need to set BENCHMARK_LENGTH to -1 to
>> > suppress these separate benchmark (there will be just one then: Raw).
>> I tested bcrypt and scrypt and the same situation
>> none@...e ~/Desktop/r/run $ ./john --test --format=bcrypt
>> Will run 8 OpenMP threads
>> Benchmarking: bcrypt ("$2a$05", 32 iterations) [Blowfish 32/64 X2]...
>> (8xOMP) using different password for benchmarking
>> Speed for cost 1 (iteration count) of 32
>> Many salts:     6096 c/s real, 768 c/s virtual
>> Only one salt:  6225 c/s real, 782 c/s virtual
> This might be CPU clock frequency scaling being slow to kick in (un-idle
> your CPU), or other activity on the system being slow to calm down after
> you've just started john.  You may try running longer benchmarks, e.g.
> use "--test=10".

I'm testing bcrypt,argon2i and argon2d in well now
in bcrypt many salts sometimes is faster, sometimes is slower
now argon2d many salts is always faster than only one (and argon2d is
faster now, many salts: 8008 l Only one salt: 7960)
argon2i many salts is slower, faster or equal

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.