john-users - Compiling with Cygwin64 in Win10: OpenCL library not found

Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [day] [month] [year] [list]
Date: Sat, 26 Dec 2015 00:45:41 +0000 (UTC)
From: - - <stadtpirat11@...il.com>
To: "john-users@...ts.openwall.com" <john-users@...ts.openwall.com>
Subject: Compiling with Cygwin64 in Win10: OpenCL library not found

Hi,

this is the first time using and compiling JTR, aswell as using cygwin64.

I haven't found any binaries with pfx2john.exe and John.exe that supports PFX/PKCS12 that also runs with CUDA enabled. So I have to compile it myself.

I mostly followed those explanations:
 - http://openwall.info/wiki/john/tutorials/win64-howto-build
 - http://openwall.info/wiki/john/GPU

In .bashrc I added the following lines:
 - export GCC=gcc
 - export PATH=$PATH:/usr/local/cuda/bin
 - export LD_LIBRARY_PATH=:/usr/local/cuda/lib64

Beside that I had to link the libraries as following, or it would not even find the OpenCL headers. I don't know, if that approach is right or wrong, or could be done in a better way.

 - ln -s /usr/local/cuda/include/CL /usr/include/
 - ln -s /usr/local/cuda/include/CL /usr/include/OpenCL


But I am having a problem with compiling JTR with CUDA enabled. I installed the CUDA SDK and the header files are found. But for some reason, the library is not found. Could someone help me out, or point me to the right direction?

Thank you!



This is the output of ./configure:
-----------------------------------------------------------------
$ ./configure --enable-cuda
[...]
checking if more special flags are required for pthreads... no
checking for PTHREAD_PRIO_INHERIT... no
checking whether we are using the Microsoft C compiler... no
checking CL/cl.h usability... yes
checking CL/cl.h presence... yes
checking for CL/cl.h... yes
checking OpenCL/cl.h usability... yes
checking OpenCL/cl.h presence... yes
checking for OpenCL/cl.h... yes
checking windows.h usability... yes
checking windows.h presence... yes
checking for windows.h... yes
checking for OpenCL library... no
checking additional paths for CUDA...  -L/usr/local/cuda/lib64 -I/usr/local/cuda/include
checking for nvcc... /cygdrive/c/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v7.5/bin/nvcc
checking for llvm-gcc-4.2... no
checking for gcc-4.6... no
checking for gcc-4.5... no
checking for gcc-4.4... no
checking for gcc-4.3... no
checking for gcc-4.2... no
checking cuda.h usability... yes
checking cuda.h presence... yes
checking for cuda.h... yes
checking for cudaGetDeviceCount in -lcudart... no
configure: error: in `/home/Daniel/JohnTheRipper/src':
configure: error: Could not find all required CUDA components
See `config.log' for more details
-----------------------------------------------------------------



The config.log shows:
-----------------------------------------------------------------
configure:11145: checking for OpenCL library
configure:11188: gcc -o conftest.exe -g -O2    -L/usr/local/lib conftest.c -lOpenCL  -lm -lkernel32  >&5
/usr/lib/gcc/x86_64-pc-cygwin/4.9.3/../../../../x86_64-pc-cygwin/bin/ld: cannot find -lOpenCL
collect2: error: ld returned 1 exit status
-----------------------------------------------------------------


FULL config.log: 
http://pastebin.com/HMXQ7i3x
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.