[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5992c4a7ae78363231f2aee28168ff6d@smtp.hushmail.com>
Date: Sun, 12 Aug 2012 04:03:21 +0200
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Jumbo 7
I tried the same on bull (actually just 'for i in `../run/john | grep
-Eo "[^ ]+\-opencl"`; do ../run/john -t -fo:$i; done'). The only problem
I saw was with EncFS. Possibly it should have LWS pegged to 32 or
perhaps 64 instead of using find_best().
magnum
On 2012-08-12 01:27, Lukas Odzioba wrote:
> 2012/8/11 Lukas Odzioba <lukas.odzioba@...il.com>:
>>> On 2012-08-10, at 5:24 PM, magnum <john.magnum@...hmail.com> wrote:
>>>
>>> All, please test and report any problems. I am not aware of any problems.
>>
>> Bad news?, during:
>
> I used this to check opencl formats:
> #!/bin/bash
> formats=(agilekeychain-opencl bf-opencl encfs-opencl keychain-opencl
> md5crypt-opencl mscash2-opencl mysql-sha1-opencl nt-opencl odf-opencl
> phpass-opencl pwsafe-opencl raw-md4-opencl raw-md5-opencl
> raw-sha1-opencl raw-sha512-opencl sha512crypt-opencl ssha-opencl
> sxc-opencl wpapsk-opencl zip-opencl xsha512-opencl )
> dir="magnum-jumbo_pwsafe2/run/"
> for fmt in ${formats[*]}
> do
> time $dir/john -test -fo=$fmt
> done
>
> 1) A lot failed formats.
> 2) I had to kill : EncFS PBKDF2 AES because of long time execution
> 3) bf and sha512crypt causes X hang - reboot needed
>
> Complete logs for OpenCL formats:
> ---------------------------------------------------------------------GTX460:
> ukasz@...kstar:~/JtR/norem_pwsafe2$ ./jtr_bencher.sh
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Benchmarking: 1Password Agile Keychain PBKDF2-HMAC-SHA-1 AES [OpenCL]... DONE
> Raw: 47786 c/s real, 48505 c/s virtual
>
>
> real 0m12.358s
> user 0m8.089s
> sys 0m4.048s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'blowfish' for 'sm_21'
> ptxas info : Function properties for blowfish
> 72 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 62 registers, 4+0 bytes lmem, 32768+0 bytes smem,
> 60 bytes cmem[0]
> ****Please see 'opencl_bf_std.h' for device specific optimizations****
> Benchmarking: OpenBSD Blowfish (x32) [OpenCL]... DONE
> Raw: 412 c/s real, 412 c/s virtual
>
>
> real 2m0.419s
> user 0m23.737s
> sys 1m36.145s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Benchmarking: EncFS PBKDF2 AES / Blowfish [OpenCL]...
> ./jtr_bencher.sh: line 6: 26462 Killed $dir/john
> -test -fo=$fmt
>
> real 17m28.128s
> user 0m4.100s
> sys 17m19.311s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Benchmarking: Mac OS X Keychain PBKDF2-HMAC-SHA-1 3DES [OpenCL]... DONE
> Raw: 56888 c/s real, 57123 c/s virtual
>
>
> real 0m12.730s
> user 0m7.932s
> sys 0m4.685s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'cryptmd5' for 'sm_21'
> ptxas info : Function properties for cryptmd5
> 272 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 35 registers, 4+0 bytes lmem, 44 bytes cmem[0],
> 11 bytes cmem[2], 8 bytes cmem[16]
> Benchmarking: md5crypt [OpenCL]... DONE
> Raw: 129196 c/s real, 130415 c/s virtual
>
>
> real 0m8.703s
> user 0m4.610s
> sys 0m3.974s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'PBKDF2' for 'sm_21'
> ptxas info : Function properties for PBKDF2
> 88 bytes stack frame, 20 bytes spill stores, 20 bytes spill loads
> ptxas info : Used 63 registers, 4+0 bytes lmem, 160+0 bytes smem,
> 52 bytes cmem[0]
> Optimal Work Group Size:128
> Kernel Execution Speed (Higher is better):0.217905
> Benchmarking: M$ Cache Hash 2 (DCC2) PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw: 14022 c/s real, 14086 c/s virtual
>
>
> real 2m20.884s
> user 0m23.933s
> sys 1m56.348s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Local work size (LWS) 256, Global work size (GWS) 2097152
> Benchmarking: MySQL 4.1 double-SHA-1 [OpenCL]... FAILED
> (get_hash[0](0))
>
> real 0m8.546s
> user 0m4.561s
> sys 0m3.844s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'nt_crypt' for 'sm_21'
> ptxas info : Function properties for nt_crypt
> 48 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 23 registers, 4+0 bytes lmem, 44 bytes cmem[0]
> Benchmarking: NT MD4 [OpenCL]... DONE
> Raw: 34499K c/s real, 35044K c/s virtual
>
>
> real 0m3.612s
> user 0m2.472s
> sys 0m0.981s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'odf' for 'sm_21'
> ptxas info : Function properties for odf
> 368 bytes stack frame, 84 bytes spill stores, 88 bytes spill loads
> ptxas info : Used 63 registers, 4+0 bytes lmem, 44 bytes cmem[0]
> Benchmarking: ODF SHA-1 Blowfish / SHA-256 AES [OpenCL]... DONE
> Raw: 11616 c/s real, 11616 c/s virtual
>
>
> real 0m24.505s
> user 0m18.103s
> sys 0m6.216s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Benchmarking: phpass MD5 ($P$9 length 8) [OpenCL]... DONE
> Raw: 267005 c/s real, 269524 c/s virtual
>
>
> real 0m6.347s
> user 0m3.282s
> sys 0m2.925s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'pwsafe' for 'sm_21'
> ptxas info : Function properties for pwsafe
> 256 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 37 registers, 4+0 bytes lmem, 44 bytes cmem[0],
> 311 bytes cmem[2]
> Benchmarking: Password Safe SHA-256 [OpenCL]... DONE
> Raw: 47003 c/s real, 47261 c/s virtual
>
>
> real 0m26.464s
> user 0m13.150s
> sys 0m13.015s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Local work size (LWS) 128, Global work size (GWS) 2097152
> Benchmarking: Raw MD4 [OpenCL]... DONE
> Raw: 21112K c/s real, 21112K c/s virtual
>
>
> real 0m8.515s
> user 0m4.720s
> sys 0m3.674s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'md5' for 'sm_21'
> ptxas info : Function properties for md5
> 64 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 27 registers, 4+0 bytes lmem, 44 bytes cmem[0]
> Local work size (LWS) 512, Global work size (GWS) 2097152
> Benchmarking: Raw MD5 [OpenCL]... DONE
> Raw: 22795K c/s real, 22961K c/s virtual
>
>
> real 0m8.484s
> user 0m4.682s
> sys 0m3.682s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Local work size (LWS) 256, Global work size (GWS) 2097152
> Benchmarking: Raw SHA-1 OpenCL [OpenCL]... FAILED (get_hash[1](0))
>
> real 0m4.886s
> user 0m2.629s
> sys 0m2.133s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'kernel_cmp' for 'sm_21'
> ptxas info : Function properties for kernel_cmp
> 0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 8 registers, 44 bytes cmem[0], 681 bytes cmem[2]
> ptxas info : Compiling entry function 'kernel_sha512' for 'sm_21'
> ptxas info : Function properties for kernel_sha512
> 200 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 61 registers, 4+0 bytes lmem, 40 bytes cmem[0],
> 681 bytes cmem[2]
> Global work size = 524288
> Benchmarking: Raw SHA-512 [OpenCL]... FAILED (get_hash[0](0))
>
> real 0m2.820s
> user 0m2.112s
> sys 0m0.598s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Building the kernel, this could take a while
> ./jtr_bencher.sh: line 6: 27532 Segmentation fault $dir/john -test -fo=$fmt
>
> real 0m25.176s
> user 0m24.541s
> sys 0m0.309s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'sha1_crypt_kernel' for 'sm_21'
> ptxas info : Function properties for sha1_crypt_kernel
> 64 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 31 registers, 4+0 bytes lmem, 48 bytes cmem[0]
> Local work size (LWS) 256, Global work size (GWS) 4194304
> Benchmarking: Netscape LDAP salted SHA-1 [OpenCL]... FAILED (get_hash[0](0))
>
> real 0m8.910s
> user 0m4.534s
> sys 0m4.206s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Benchmarking: SXC SHA-1 Blowfish [OpenCL]... DONE
> Raw: 12517 c/s real, 12624 c/s virtual
>
>
> real 1m16.909s
> user 0m52.963s
> sys 0m23.233s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'wpapsk' for 'sm_21'
> ptxas info : Function properties for wpapsk
> 312 bytes stack frame, 40 bytes spill stores, 28 bytes spill loads
> ptxas info : Used 63 registers, 4+0 bytes lmem, 44 bytes cmem[0]
> Benchmarking: WPA-PSK PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw: 13824 c/s real, 14034 c/s virtual
>
>
> real 0m30.324s
> user 0m18.114s
> sys 0m11.868s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Benchmarking: ZIP-AES PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw: 33253 c/s real, 33425 c/s virtual
>
>
> real 0m16.090s
> user 0m9.422s
> sys 0m6.470s
> OpenCL platform 0: NVIDIA CUDA, 1 device(s).
> Using device 0: GeForce GTX 460
> Compilation log:
> ptxas info : Compiling entry function 'kernel_cmp' for 'sm_21'
> ptxas info : Function properties for kernel_cmp
> 0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 8 registers, 44 bytes cmem[0], 681 bytes cmem[2]
> ptxas info : Compiling entry function 'kernel_xsha512' for 'sm_21'
> ptxas info : Function properties for kernel_xsha512
> 200 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
> ptxas info : Used 62 registers, 4+0 bytes lmem, 44 bytes cmem[0],
> 681 bytes cmem[2]
> Local work size (LWS) 128, Global work size (GWS) 524288
> Benchmarking: Mac OS X 10.7+ salted SHA-512 [OpenCL]... FAILED (get_hash[0](0))
>
> real 0m2.879s
> user 0m2.078s
> sys 0m0.612s
> ---------------------------------------------------------------------5850:
> ukasz@...kstar:~/JtR/norem_pwsafe2$ ./jtr_bencher.sh
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: 1Password Agile Keychain PBKDF2-HMAC-SHA-1 AES [OpenCL]... DONE
> Raw: 79448 c/s real, 81317 c/s virtual
>
>
> real 0m14.910s
> user 0m9.956s
> sys 0m4.535s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: EncFS PBKDF2 AES / Blowfish [OpenCL]...
> ./jtr_bencher.sh: line 7: 2823 Killed $dir/john
> -test -plat=1 -dev=0 -fo=$fmt
>
> real 9m9.670s
> user 0m0.549s
> sys 8m48.473s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: Mac OS X Keychain PBKDF2-HMAC-SHA-1 3DES [OpenCL]... DONE
> Raw: 108000 c/s real, 108850 c/s virtual
>
>
> real 0m15.207s
> user 0m8.564s
> sys 0m4.478s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: md5crypt [OpenCL]... DONE
> Raw: 119172 c/s real, 119172 c/s virtual
>
>
> real 0m8.330s
> user 0m2.951s
> sys 0m5.210s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Optimal Work Group Size:256
> Kernel Execution Speed (Higher is better):0.502145
> Benchmarking: M$ Cache Hash 2 (DCC2) PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw: 32105 c/s real, 16394 c/s virtual
>
>
> real 1m2.639s
> user 0m39.412s
> sys 0m56.991s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Local work size (LWS) 128, Global work size (GWS) 2097152
> Benchmarking: MySQL 4.1 double-SHA-1 [OpenCL]... DONE
> Many salts: 14364K c/s real, 15887K c/s virtual
> Only one salt: 13981K c/s real, 14979K c/s virtual
>
>
> real 0m20.771s
> user 0m4.370s
> sys 0m12.796s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: NT MD4 [OpenCL]... DONE
> Raw: 30583K c/s real, 31457K c/s virtual
>
>
> real 0m2.896s
> user 0m1.430s
> sys 0m1.301s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: ODF SHA-1 Blowfish / SHA-256 AES [OpenCL]... DONE
> Raw: 15191 c/s real, 15360 c/s virtual
>
>
> real 0m25.713s
> user 0m19.301s
> sys 0m5.967s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: phpass MD5 ($P$9 length 8) [OpenCL]... DONE
> Raw: 274707 c/s real, 274707 c/s virtual
>
>
> real 0m4.928s
> user 0m1.330s
> sys 0m3.486s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Compilation log: /tmp/OCLYpaNHb.cl(166): warning: a value of type
> "__global uchar *" cannot be
> used to initialize an entity of type "__global uint *"
> __global uint32_t *v = salt->hash;
> ^
>
> LOOP UNROLL: pragma unroll (line 151)
> Unrolled as requested!
> LOOP UNROLL: pragma unroll (line 112)
> Unrolled as requested!
>
> Benchmarking: Password Safe SHA-256 [OpenCL]... DONE
> Raw: 34683 c/s real, 34683 c/s virtual
>
>
> real 0m32.221s
> user 0m2.663s
> sys 0m29.342s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Local work size (LWS) 128, Global work size (GWS) 2097152
> Benchmarking: Raw MD4 [OpenCL]... DONE
> Raw: 19418K c/s real, 19538K c/s virtual
>
>
> real 0m7.544s
> user 0m1.801s
> sys 0m5.625s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Local work size (LWS) 128, Global work size (GWS) 2097152
> Benchmarking: Raw MD5 [OpenCL]... DONE
> Raw: 21112K c/s real, 21399K c/s virtual
>
>
> real 0m7.602s
> user 0m1.778s
> sys 0m5.683s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Local work size (LWS) 128, Global work size (GWS) 2097152
> Benchmarking: Raw SHA-1 OpenCL [OpenCL]... DONE
> Raw: 25368K c/s real, 25575K c/s virtual
>
>
> real 0m8.312s
> user 0m2.572s
> sys 0m5.574s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Compilation log: LOOP UNROLL: pragma unroll (line 163)
> Unrolled as requested!
> LOOP UNROLL: pragma unroll (line 168)
> Unrolled as requested!
> LOOP UNROLL: pragma unroll (line 183)
> Unrolled as requested!
>
> Global work size = 524288
> Benchmarking: Raw SHA-512 [OpenCL]... DONE
> Raw: 15049K c/s real, 15049K c/s virtual
>
>
> real 0m5.762s
> user 0m3.325s
> sys 0m2.228s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Local work size (LWS) 128, Global work size (GWS) 4194304
> Benchmarking: Netscape LDAP salted SHA-1 [OpenCL]... DONE
> Many salts: 34568K c/s real, 34759K c/s virtual
> Only one salt: 24672K c/s real, 24672K c/s virtual
>
>
> real 0m20.800s
> user 0m2.583s
> sys 0m17.949s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: SXC SHA-1 Blowfish [OpenCL]... DONE
> Raw: 15392 c/s real, 15489 c/s virtual
>
>
> real 0m46.358s
> user 0m26.738s
> sys 0m19.126s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: WPA-PSK PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw: 27806 c/s real, 27927 c/s virtual
>
>
> real 0m20.215s
> user 0m10.956s
> sys 0m8.955s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Benchmarking: ZIP-AES PBKDF2-HMAC-SHA-1 [OpenCL]... DONE
> Raw: 76375 c/s real, 79448 c/s virtual
>
>
> real 0m15.431s
> user 0m7.914s
> sys 0m7.172s
> OpenCL platform 1: AMD Accelerated Parallel Processing, 2 device(s).
> Using device 0: Cypress
> Compilation log: LOOP UNROLL: pragma unroll (line 163)
> Unrolled as requested!
> LOOP UNROLL: pragma unroll (line 168)
> Unrolled as requested!
> LOOP UNROLL: pragma unroll (line 183)
> Unrolled as requested!
>
> Local work size (LWS) 256, Global work size (GWS) 524288
> Benchmarking: Mac OS X 10.7+ salted SHA-512 [OpenCL]... DONE
> Many salts: 35606K c/s real, 35606K c/s virtual
> Only one salt: 17476K c/s real, 13107K c/s virtual
>
>
> real 0m7.683s
> user 0m3.094s
> sys 0m4.306s
>
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.
