[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sun, 8 Jul 2012 10:19:38 +0200
From: Frank Dittrich <frank_dittrich@...mail.com>
To: john-dev@...ts.openwall.com
Subject: Re: -atom make targets
On 07/08/2012 08:44 AM, Solar Designer wrote:
> On Sat, Jul 07, 2012 at 11:49:48AM +0200, Frank Dittrich wrote:
>> Benchmarking: FreeBSD MD5 [32/32]... 3157 c/s real, 3157 c/s virtual
>> Benchmarking: FreeBSD MD5 [32/32 X2]... 2720 c/s real, 2720 c/s virtual
>> Benchmarking: FreeBSD MD5 [32/32]... 2360 c/s real, 2360 c/s virtual
Since the primary focus here was the build error, I am not sure whether
I made sure nothing else was running in parallel.
So for this mail, I was a bit more careful.
> I think this suggests that MD5_IMM = 0 is a better choice for Atom.
> This is totally unexpected to me. MD5_IMM = 0 is normally for RISC
> archs, not for x86.
>
> Frank, do you also get a speedup for MD5 with generic over -x86- builds,
> if you try non-jumbo or if you try -x86-any with jumbo? If so, as Jim
> explained, this matters for some dynamic formats in jumbo as well (even
> though you obviously get much higher speeds for --format=md5 in jumbo
> due to SSE2).
Here is the output of building generic on john-1.7.9 (core), OS Fedora
16, Linux kernel version 3.4.2-1.fc16.i686.PAE #1 SMP, gcc (GCC) 4.6.3
20120306 (Red Hat 4.6.3-2):
$ make clean generic > /dev/null
Benchmarking: Traditional DES [24/32 4K]... 57292 c/s real, 57407 c/s
virtual
Benchmarking: Traditional DES [24/32 128K]... 51046 c/s real, 51148 c/s
virtual
Benchmarking: Traditional DES [24/32 4K]... 58496 c/s real, 58613 c/s
virtual
Benchmarking: Traditional DES [24/32 4K]... 74649 c/s real, 74799 c/s
virtual
Benchmarking: Traditional DES [24/32 4K]... 71014 c/s real, 71156 c/s
virtual
Benchmarking: Traditional DES [32/32 BS]... 170201 c/s real, 170542 c/s
virtual
Benchmarking: Traditional DES [32/32 BS]... 179353 c/s real, 180073 c/s
virtual
Benchmarking: Traditional DES [32/32 BS]... 105881 c/s real, 106093 c/s
virtual
Benchmarking: FreeBSD MD5 [32/32]... 3201 c/s real, 3207 c/s virtual
Benchmarking: FreeBSD MD5 [32/32 X2]... 2796 c/s real, 2802 c/s virtual
Benchmarking: FreeBSD MD5 [32/32]... 3200 c/s real, 3206 c/s virtual
Benchmarking: OpenBSD Blowfish (x32) [32/32]... 186 c/s real, 187 c/s
virtual
Benchmarking: OpenBSD Blowfish (x32) [32/32]... 220 c/s real, 220 c/s
virtual
Benchmarking: OpenBSD Blowfish (x32) [32/32 X2]... 170 c/s real, 170 c/s
virtual
[fd@fdn run]$ ./john-linux-x86-sse2 --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 4255 c/s real, 4264 c/s virtual
[fd@fdn run]$ ./john-linux-x86-sse2 --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 4298 c/s real, 4307 c/s virtual
[fd@fdn run]$ ./john-linux-x86-mmx --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 4296 c/s real, 4300 c/s virtual
[fd@fdn run]$ ./john-linux-x86-mmx --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 4274 c/s real, 4278 c/s virtual
[fd@fdn run]$ ./john-linux-x86-any --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 4272 c/s real, 4281 c/s virtual
[fd@fdn run]$ ./john-linux-x86-any --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 4267 c/s real, 4275 c/s virtual
[fd@fdn run]$ ./john-generic --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 3191 c/s real, 3198 c/s virtual
[fd@fdn run]$ ./john-generic --test=10 --format=md5
Benchmarking: FreeBSD MD5 [32/32]... DONE
Raw: 3170 c/s real, 3176 c/s virtual
So, sse, mmx and any are similar, generic is disappointing.
That looks like there is something wrong with build generic (or with
bench.c).
For latest git, I'll just compare sse2, any and generic.
But I'll have to postpone it until later today.
Frank
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.
