[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 11 Apr 2015 02:35:05 +0200
From: Agnieszka Bielec <bielecagnieszka8@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: [GSoC] John the Ripper support for PHC finalists
2015-04-11 2:11 GMT+02:00 Solar Designer <solar@...nwall.com>:
> On Sat, Apr 11, 2015 at 01:54:47AM +0200, Agnieszka Bielec wrote:
>> 2015-04-11 1:34 GMT+02:00 Solar Designer <solar@...nwall.com>:
>> >> I've added SSE2 and isn't faster (bleeding-jumbo)
>> >
>> > This is unexpected. Are you sure the SSE2 (actually AVX, when building
>> > with AVX enabled) code is getting compiled in? And the non-SSE2 code
>> > isn't getting compiled in?
>>
>> I put printf() into POMELO_SSE2()
>
> or/and forgetting to set GOMP_CPU_AFFINITY=0-31
I don't know, when I type GOMP_CPU_AFFINITY=x before ./john I have even worse
results
> Can you please post shell commands & output of how you benchmark the old
> non-SSE2 vs. the new SSE2 code?
SSE2
[a@...er run]$ ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 21291 c/s real, 690 c/s virtual
Only one salt: 20480 c/s real, 661 c/s virtual
[a@...er run]$ OMP_SCALE=2 ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 45176 c/s real, 1467 c/s virtual
Only one salt: 44609 c/s real, 1440 c/s virtual
[a@...er run]$ OMP_SCALE=4 ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 68266 c/s real, 2306 c/s virtual
Only one salt: 77053 c/s real, 2564 c/s virtual
[a@...er run]$ OMP_SCALE=8 ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 89219 c/s real, 3040 c/s virtual
Only one salt: 86646 c/s real, 2904 c/s virtual
non-SSE
[a@...er run]$ ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) form
DONE
Many salts: 23090 c/s real, 766 c/s virtual
Only one salt: 26101 c/s real, 856 c/s virtual
[a@...er run]$ OMP_SCALE=2 ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) form
DONE
Many salts: 44032 c/s real, 1422 c/s virtual
Only one salt: 42749 c/s real, 1390 c/s virtual
[a@...er run]$ OMP_SCALE=4 ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) form
DONE
Many salts: 64250 c/s real, 2069 c/s virtual
Only one salt: 66258 c/s real, 2136 c/s virtual
[a@...er run]$ OMP_SCALE=8 ./john --format=pomelo --test
Will run 32 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (32xOMP) form
DONE
Many salts: 76298 c/s real, 2687 c/s virtual
Only one salt: 79533 c/s real, 2671 c/s virtual
[a@...er run]$ OMP_NUM_THREADS=1 ./john --format=pomelo --test
Warning: OpenMP is disabled; a non-OpenMP build may be faster
Benchmarking: pomelo, Generic pomelo [Pomelo]... form
DONE
Many salts: 10928 c/s real, 10928 c/s virtual
Only one salt: 10928 c/s real, 10928 c/s virtual
[a@...er run]$ OMP_NUM_THREADS=2 ./john --format=pomelo --test
Will run 2 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (2xOMP) form
DONE
Many salts: 21760 c/s real, 10880 c/s virtual
Only one salt: 21472 c/s real, 10789 c/s virtual
[a@...er run]$ OMP_NUM_THREADS=4 ./john --format=pomelo --test
Will run 4 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (4xOMP) form
DONE
Many salts: 41664 c/s real, 10442 c/s virtual
Only one salt: 42176 c/s real, 10570 c/s virtual
[a@...er run]$ OMP_NUM_THREADS=8 ./john --format=pomelo --test
Will run 8 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (8xOMP) form
DONE
Many salts: 81408 c/s real, 10201 c/s virtual
Only one salt: 81408 c/s real, 10176 c/s virtual
SSE2
[a@...er run]$ OMP_NUM_THREADS=1 ./john --format=pomelo --test
Warning: OpenMP is disabled; a non-OpenMP build may be faster
Benchmarking: pomelo, Generic pomelo [Pomelo]... DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 12992 c/s real, 12992 c/s virtual
Only one salt: 12992 c/s real, 12992 c/s virtual
[a@...er run]$ OMP_NUM_THREADS=2 ./john --format=pomelo --test
Will run 2 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (2xOMP) DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 25600 c/s real, 12864 c/s virtual
Only one salt: 25856 c/s real, 12928 c/s virtual
[a@...er run]$ OMP_NUM_THREADS=4 ./john --format=pomelo --test
Will run 4 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (4xOMP) DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 50176 c/s real, 12575 c/s virtual
Only one salt: 50112 c/s real, 12528 c/s virtual
[a@...er run]$ OMP_NUM_THREADS=8 ./john --format=pomelo --test
Will run 8 OpenMP threads
Benchmarking: pomelo, Generic pomelo [Pomelo]... (8xOMP) DONE
Speed for cost 1 (N) of 2, cost 2 (r) of 2
Many salts: 96256 c/s real, 12062 c/s virtual
Only one salt: 95683 c/s real, 12080 c/s virtual
It is slightly faster
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.
