Closed GoogleCodeExporter closed 8 years ago
please give information about gcc version. And version of CAL++ library.
( output of command gcc -v )
Original comment by hazema...@gmail.com
on 15 Apr 2010 at 12:50
Also please give version of boost library on your system and verify if it's
installed
correctly.
It looks a little bit like boost::is_same is missing ( maybe some additional
includes
are required ).
Original comment by hazema...@gmail.com
on 15 Apr 2010 at 1:09
root@bt:~# gcc -v
Using built-in specs.
Target: i486-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.3.2-1ubuntu12'
--with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs
--enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared
--with-system-zlib --libexecdir=/usr/lib --without-included-gettext
--enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.3
--program-suffix=-4.3 --enable-clocale=gnu --enable-libstdcxx-debug
--enable-objc-gc
--enable-mpfr --enable-targets=all --enable-checking=release
--build=i486-linux-gnu
--host=i486-linux-gnu --target=i486-linux-gnu
Thread model: posix
gcc version 4.3.2 (Ubuntu 4.3.2-1ubuntu12)
CAL++ library is calpp-0.86.3.tar.gz
and boost is from ubuntu packet manager (look to attached file)
Original comment by odl...@gmail.com
on 15 Apr 2010 at 2:06
Attachments:
I've solved...removed all libbost library from packet manager and reinstall
from source
This is the benchmark
pyrit benchmark
Pyrit 0.3.1-dev (svn r247) (C) 2008-2010 Lukas Lueg http://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3+
Running benchmark (77958.0 PMKs/s)... /
Computed 81288.19 PMKs/s total.
#1: 'CAL++ Device #1 'ATI CYPRESS'': 40546.6 PMKs/s (RTT 2.8)
#2: 'CAL++ Device #2 'ATI CYPRESS'': 39239.6 PMKs/s (RTT 2.7)
#3: 'CPU-Core (SSE2)': 457.6 PMKs/s (RTT 3.1)
#4: 'CPU-Core (SSE2)': 460.1 PMKs/s (RTT 3.0)
#5: 'CPU-Core (SSE2)': 463.7 PMKs/s (RTT 2.8)
#6: 'CPU-Core (SSE2)': 455.8 PMKs/s (RTT 2.9)
#7: 'CPU-Core (SSE2)': 470.4 PMKs/s (RTT 3.0)
#8: 'CPU-Core (SSE2)': 466.0 PMKs/s (RTT 3.1)
Original comment by odl...@gmail.com
on 15 Apr 2010 at 5:03
fixed. this is a stock 5970 ?
Original comment by lukas.l...@gmail.com
on 15 Apr 2010 at 5:55
SAPPHIRE RADEON HD 5970 OverClock edition but currently use default value.
Any benefit if I switch to 64bit system?
Original comment by odl...@gmail.com
on 15 Apr 2010 at 6:06
Hard to say. What drivers version are you using ?
To tell you the truth on ati forum there was report 45 pmks/s on 5850.
And each core in 5970 should be slightly faster then 5850. Strange.
I have a hunch that the card is not feeded properly.
You could try to change line 410 in pyrit/cpyrit/cpyrit.py from 'ncpus-=1' to
'ncpus-=2' ( or even 4 ). And please report the results :).
If it will help then it's probably a "bug" in pyrit.
And the problem was with 1.34 version of boost library - it's quite old :).
There were some additional includes required. But I'll fix it anyway in CAL++
svn soon.
Original comment by hazema...@gmail.com
on 15 Apr 2010 at 6:44
I really would like you to try change this line with ncpus. It most probably is
a
problem in pyrit that needs to be addressed. On nathalem each core is reported
as 2 (
with hyperthreading enabled ). And this leads to one real core serving 2 gpus.
The
problem is also magnified by busy waiting in cal drivers.
The effect could be that both gpus are starving and don't run on full speed.
Original comment by hazema...@gmail.com
on 15 Apr 2010 at 7:24
I use ati driver 10.3
change line 410 in /usr/lib/python2.5/site-packages/cpyrit/cpyrit.py from
'ncpus-=1'
to ncpus-=2
pyrit benchmark
Pyrit 0.3.1-dev (svn r247) (C) 2008-2010 Lukas Lueg http://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3+
Running benchmark (83321.4 PMKs/s)... -
Computed 89267.17 PMKs/s total.
#1: 'CAL++ Device #1 'ATI CYPRESS'': 45736.9 PMKs/s (RTT 2.7)
#2: 'CAL++ Device #2 'ATI CYPRESS'': 41201.6 PMKs/s (RTT 2.8)
#3: 'CPU-Core (SSE2)': 561.0 PMKs/s (RTT 2.9)
#4: 'CPU-Core (SSE2)': 624.2 PMKs/s (RTT 3.0)
#5: 'CPU-Core (SSE2)': 595.8 PMKs/s (RTT 2.9)
#6: 'CPU-Core (SSE2)': 536.2 PMKs/s (RTT 3.0)
change line 410 in /usr/lib/python2.5/site-packages/cpyrit/cpyrit.py from
'ncpus-=2'
to ncpus-=4
pyrit benchmark
Pyrit 0.3.1-dev (svn r247) (C) 2008-2010 Lukas Lueg http://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3+
Running benchmark (90389.9 PMKs/s)... /
Computed 93075.44 PMKs/s total.
#1: 'CAL++ Device #1 'ATI CYPRESS'': 47712.3 PMKs/s (RTT 2.8)
#2: 'CAL++ Device #2 'ATI CYPRESS'': 47148.3 PMKs/s (RTT 2.8)
no more CPU-Core ??
Original comment by odl...@gmail.com
on 16 Apr 2010 at 7:24
I would say typical trade-off. You give back 2800 pmk from cpus for 15000 pmk
from
gpu. I think it's a good deal :).
The whole problem is that ATI drivers are really cpu sensitive ( or cpu hogs :)
).
I'm working on new method for feeding gpu. It's done and running, but it
requires
more testing. The idea was to be less cpu sensitive and have 1% maybe 2%
speedup from
gpu. The good thing is that cpu usage dropped. But the strange thing is that I
have
9% speedup from gpu. I've tested it and everything looks ok, but because of the
9% I
am not sure. After some cleanups I'll probably post it as issue here so people
can
test it a little. And then Lukas will decide what to do with it.
Also CAL++ has experimental support for non blocking waiting for events. It's
disabled by default. To enable it you need to uncomment line 43 in cal.hpp ( or
add
-D__CAL_USE_NON_BLOCKING_WAIT to gcc parameters ). It has it's own issues, but
maybe
it will help a little bit.
Original comment by hazema...@gmail.com
on 16 Apr 2010 at 9:23
This after uncomment line 42 in /usr/local/include/cal/cal.hpp
#define __CAL_USE_NON_BLOCKING_WAIT 1
root@bt:~# pyrit benchmark
Pyrit 0.3.1-dev (svn r247) (C) 2008-2010 Lukas Lueg http://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3+
Running benchmark (89656.3 PMKs/s)... /
Computed 94369.22 PMKs/s total.
#1: 'CAL++ Device #1 'ATI CYPRESS'': 47348.0 PMKs/s (RTT 2.8)
#2: 'CAL++ Device #2 'ATI CYPRESS'': 48230.1 PMKs/s (RTT 2.9)
root@bt:~# pyrit benchmark
Pyrit 0.3.1-dev (svn r247) (C) 2008-2010 Lukas Lueg http://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3+
Running benchmark (90406.6 PMKs/s)... -
Computed 94281.00 PMKs/s total.
#1: 'CAL++ Device #1 'ATI CYPRESS'': 47893.6 PMKs/s (RTT 2.7)
#2: 'CAL++ Device #2 'ATI CYPRESS'': 46314.2 PMKs/s (RTT 2.8)
Original comment by odl...@gmail.com
on 16 Apr 2010 at 10:11
I'm trying to install cpyrit_calpp under Ubuntu 64bit but when
odlan@H3ll:~/Scaricati/pyrit/pyrit_svn/cpyrit_calpp$ sudo python setup.py
install
unavailable enviroment variable ATISTREAMSDKROOT
Traceback (most recent call last):
File "setup.py", line 32, in <module>
CALPP_INC_DIR = os.environ['ATISTREAMSDKROOT']
File "/usr/lib/python2.6/UserDict.py", line 22, in __getitem__
raise KeyError(key)
KeyError: 'ATISTREAMSDKROOT'
odlan@H3ll:~/Scaricati/pyrit/pyrit_svn/cpyrit_calpp$ echo $ATISTREAMSDKROOT
/home/odlan/Scaricati/ati-stream-sdk-v2.01-lnx64/
odlan@H3ll:~$ python --version
Python 2.6.4
Original comment by odl...@gmail.com
on 16 Apr 2010 at 2:25
ATISTREAMSDKROOT must be pointing to directory with ATI's SDK..
As written in README and in ATI's SDK README :).
before building issue command
export ATISTREAMSDKROOT=directory with sdk
Original comment by hazema...@gmail.com
on 16 Apr 2010 at 2:32
Sudo removes some enviroment variables. So
sudo bash
export ATISTREAMSDKROOT=...
python setup.py install
Original comment by hazema...@gmail.com
on 16 Apr 2010 at 2:36
Thank you.Now fixed
pyrit benchmark
Pyrit 0.3.1-dev (C) 2008-2010 Lukas Lueg http://pyrit.googlecode.com
This code is distributed under the GNU General Public License v3+
Running benchmark (95467.3 PMKs/s)... -
Computed 100756.52 PMKs/s total.
#1: 'CAL++ Device #1 'ATI CYPRESS'': 49512.9 PMKs/s (RTT 2.8)
#2: 'CAL++ Device #2 'ATI CYPRESS'': 49408.2 PMKs/s (RTT 2.7)
no substantial improvement on 64bit system
Original comment by odl...@gmail.com
on 16 Apr 2010 at 5:48
Odlan3 could you test v2 core. I'm really interested how it will work for you.
( look at the issue 148 )
Original comment by hazema...@gmail.com
on 17 Apr 2010 at 7:30
I will try and tell you the results
Original comment by odl...@gmail.com
on 17 Apr 2010 at 8:18
[deleted comment]
I' can't compile cpyrit_calpp
odlan@H3ll:~/Scaricati/pyritV2/cpyrit_calpp$ python setup.py build
svn: '.' non è una copia locale
running build
running build_ext
Building modules...
building 'cpyrit._cpyrit_calpp' extension
creating build
creating build/temp.linux-x86_64-2.6
gcc -pthread -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall
-Wstrict-prototypes
-fPIC -I/home/odlan/Scaricati/ati-stream-sdk-v2.01-lnx64/include
-I/usr/include/python2.6 -c _cpyrit_calpp.cpp -o
build/temp.linux-x86_64-2.6/_cpyrit_calpp.o -DVERSION="0.3.1-dev"
cc1plus: warning: command line option "-Wstrict-prototypes" is valid for
Ada/C/ObjC
but not for C++
_cpyrit_calpp.cpp: In function ‘void start_kernel(CALDevice*, int)’:
_cpyrit_calpp.cpp:248: error: ‘class cal::CommandQueue’ has no member named
‘isEventDone’
_cpyrit_calpp.cpp: In function ‘void send_pmklist(CALDevice*, gpu_inbuffer*,
int)’:
_cpyrit_calpp.cpp:263: error: ‘CAL_RESALLOC_REMOTE’ was not declared in
this scope
_cpyrit_calpp.cpp:264: error: ‘CAL_RESALLOC_REMOTE’ was not declared in
this scope
_cpyrit_calpp.cpp: In function ‘PyObject* cpyrit_receive(CALDevice*,
PyObject*)’:
_cpyrit_calpp.cpp:396: error: ‘class cal::CommandQueue’ has no member named
‘isEventDone’
_cpyrit_calpp.cpp:399: warning: comparison between signed and unsigned integer
expressions
error: command 'gcc' failed with exit status 1
Original comment by odl...@gmail.com
on 17 Apr 2010 at 11:34
You need to install svn version of CAL++ library ( look at the issue 148 ). This
error is explained there.
Original comment by hazema...@gmail.com
on 17 Apr 2010 at 11:47
Hello sorry to revive this could you go over the inclusion of the
ati-stream-sdk part...
unavailable enviroment variable ATISTREAMSDKROOT
Traceback (most recent call last):
File "setup.py", line 35, in <module>
CALPP_INC_DIR = os.environ['ATISTREAMSDKROOT']
File "/usr/lib/python2.6/UserDict.py", line 22, in __getitem__
raise KeyError(key)
KeyError: 'ATISTREAMSDKROOT'
ls
build _cpyrit_calpp.cpp
_cpyrit_calpp_kernel_generator.cpp README
convert_optimized.sed _cpyrit_calpp.h generate_optimized_kernel
setup.py
COPYING _cpyrit_calpp_kernel.cpp MANIFEST.in
echo $ATISTREAMSDKROOT
/ati-stream-sdk-v2.2-lnx32
python --version
Python 2.6.6
I am going mad trying to get this to work! What commands do i need to use!
Original comment by Keylogge...@gmail.com
on 26 Jan 2011 at 12:44
For environment variable to be visible in executed programs you must use export
keyword. So you need to type
export ATISTREAMSDKROOT=/directory_with_sdk
Original comment by hazema...@gmail.com
on 26 Jan 2011 at 11:26
Issue 240 has been merged into this issue.
Original comment by lukas.l...@gmail.com
on 26 Jan 2011 at 12:05
[deleted comment]
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$ echo $ATISTREAMSDKROOT
/ati-stream-sdk-v2.2-lnx32
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$ export
ATISTREAMSDKROOT=/ati-stream-sdk-v2.2-lnx32
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$ sudo ./setup.py build
unavailable enviroment variable ATISTREAMSDKROOT
Traceback (most recent call last):
File "./setup.py", line 35, in <module>
CALPP_INC_DIR = os.environ['ATISTREAMSDKROOT']
File "/usr/lib/python2.6/UserDict.py", line 22, in __getitem__
raise KeyError(key)
KeyError: 'ATISTREAMSDKROOT'
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$
Am i missing something?
Original comment by Keylogge...@gmail.com
on 26 Jan 2011 at 2:07
You can use command
set
to see what environment variables are defined.
Or just to check if ATISTREAMSDKROOT is avaiable use
set | grep ATISTREAMSDKROOT
I don't know how to help you. Problem is that setup doesn't see this
environment variable. If it's defined maybe there is some problem with your
python installation ( try to reinstall python ).
Original comment by hazema...@gmail.com
on 26 Jan 2011 at 3:42
You may also simply replace the line
CALPP_INC_DIR = os.environ['ATISTREAMSDKROOT']
in setup.py with
CALPP_INC_DIR = '/ati-stream-sdk-v2.2-lnx32'
(while '/ati-stream-sdk-v2.2-lnx32' should in fact be the fully qualified path)
Original comment by lukas.l...@gmail.com
on 26 Jan 2011 at 5:29
[deleted comment]
please do not post such links.
if you need further help, please post on the mailing list as this bug is closed.
Original comment by lukas.l...@gmail.com
on 26 Jan 2011 at 11:03
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$ echo $ATISTREAMSDKROOT
/ati-stream-sdk-v2.2-lnx32
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$ export
ATISTREAMSDKROOT=/ati-stream-sdk-v2.2-lnx32
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$ sudo ./setup.py build
unavailable enviroment variable ATISTREAMSDKROOT
Traceback (most recent call last):
File "./setup.py", line 35, in <module>
CALPP_INC_DIR = os.environ['ATISTREAMSDKROOT']
File "/usr/lib/python2.6/UserDict.py", line 22, in __getitem__
raise KeyError(key)
KeyError: 'ATISTREAMSDKROOT'
middle@The-Beast:~/Desktop/pyrit_svn/cpyrit_calpp$
Am i missing something?
I had the same troubles using ubuntu 10 10. It was because the version of
libboost was old. Grab a copy of boost 1.46 and you should be golden ;)
Original comment by zarrens...@gmail.com
on 13 Apr 2011 at 8:36
I think this issue could be partially related to #173.
odl.., can you do not olny a benckmark but also a real test on a list.txt of
(i.e.) 500 Million of passwords and report the real PMK/s?
Please, do
time pyrit -e test -r wpa.cap -i list.txt attack_passthrough
and report result.
Original comment by pyrit.lo...@gmail.com
on 13 Apr 2011 at 10:47
Original issue reported on code.google.com by
odl...@gmail.com
on 15 Apr 2010 at 11:46Attachments: