[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Sat, 03 Dec 2011 13:26:46 -0500
From: Erik Winkler <ewinkler@...ls.com>
To: john-dev@...ts.openwall.com
Subject: Re: Use of ICC
Alexander,
Thanks, that helped, but I also had to set #define JOHN_AVX as this was not being set either.
It compiles a running binary, but I had to comment out xgetbv in x86-64.S as I was getting this error:
no such instruction: `xgetbv'
Here are some benchmarks,
./run/john -test
Benchmarking: Traditional DES [128/128 BS AVX-16]... DONEMany salts: 5616K c/s real, 5616K c/s virtual
Only one salt: 5323K c/s real, 5323K c/s virtual
Benchmarking: BSDI DES (x725) [128/128 BS AVX-16]... DONE
Many salts: 169472 c/s real, 169472 c/s virtual
Only one salt: 165580 c/s real, 165580 c/s virtual
Benchmarking: FreeBSD MD5 [32/64 X2]... DONE
Raw: 13800 c/s real, 13800 c/s virtual
Benchmarking: OpenBSD Blowfish (x32) [32/64 X2]... DONE
Raw: 844 c/s real, 846 c/s virtual
Benchmarking: Kerberos AFS DES [48/64 4K]... DONE
Short: 462592 c/s real, 462592 c/s virtual
Long: 1556K c/s real, 1556K c/s virtual
Benchmarking: LM DES [128/128 BS AVX-16]... DONE
Raw: 70491K c/s real, 70491K c/s virtual
Benchmarking: Tripcode DES [48/64 4K]... DONE
Raw: 422784 c/s real, 422784 c/s virtual
Benchmarking: dummy [N/A]... DONE
Raw: 76835K c/s real, 76835K c/s virtual
Thanks,
Erik
>
> Try removing " && defined(__GNUC__)" from this line in x86-64.h:
>
> #elif defined(JOHN_AVX) && defined(__GNUC__)
> /* Require gcc for AVX because DES_bs_all is aligned in a gcc-specific way */
>
> Don't forget a "make clean".
>
> Does this help?
>
> Thanks,
>
> Alexander
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.
