Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJXvNmvWjQtjD=m1k9q3=CzBbEqhWzX0waF3FtJoF1=Up535QA@mail.gmail.com>
Date: Tue, 11 Mar 2014 13:27:07 -0400
From: K Zug <kzug10@...il.com>
To: john-users@...ts.openwall.com
Subject: Help needed on CUDA / OpenCL on OS X 10.9

Hello All,
I have tried as far as I could, but I'll need some help/guidance on this
one:

CUDA won't compile properly, and OpenCL, even though compiling will either
hang, freeze, or return a "Warning: salt() returned misaligned pointer" or
gives a "Abort trap: 6" when working on Jeff's DMG.
Please note that when benchmarking,  the dmg-opencl benchmark correctly.

I tried as far as I could, even installed GCC 4.2 via homebrew to no avail.
I am ready to sacrifice a rooster or wear a necklace of garlic if
necessary.

Any help would be appreciated
Thanks in advance

Kris.


Config:

OS X Mavericks 10.9.2
NVIDIA GeForce GT 750M 1024 MB
CUDA 5.5
X Code Version 5.1 (5B130a)
Bleeding Jumbo, latest

gcc -v
Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr
--with-gxx-include-dir=/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.9.sdk/usr/include/c++/4.2.1
Apple LLVM version 5.1 (clang-503.0.38) (based on LLVM 3.4svn)
Target: x86_64-apple-darwin13.1.0
Thread model: posix

$ make clean macosx-x86-64-cuda

result:

cd cuda;
PATH=/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/local/cuda/bin/ nvcc
-c -Xptxas -v -fmad=true -arch sm_10 -m64 -ccbin=/usr/bin/llvm-gcc-4.2
cuda_common.cu -o ../cuda_common.o
/usr/bin/llvm-gcc-4.2: No such file or directory
make[1]: *** [cuda_common.o] Error 1
make: *** [macosx-x86-64-gpu] Error 2

--> llvm-gcc-4.2 no longer provided with OS X 10.9

changed line 83 of  src/makefile
CCBIN_OSX = -ccbin=/usr/bin/llvm-gcc-4.2
changed to
CCBIN_OSX = -ccbin=/usr/bin/llvm-gcc

result:
cd cuda;
PATH=/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/local/cuda/bin/ nvcc
-c -Xptxas -v -fmad=true -arch sm_10 -m64 -ccbin=/usr/bin/llvm-gcc
cuda_common.cu -o ../cuda_common.o
clang: error: unsupported option '-dumpspecs'
clang: error: no input files


$ make clean macosx-x86-64-opencl

Failing after this point just means some helper tools did not build:
/Applications/Xcode.app/Contents/Developer/usr/bin/make ../run/SIPdump
../run/vncpcap2john
gcc -c -Wall -Wdeclaration-after-statement -O2 -fomit-frame-pointer
-I/usr/local/include `pkg-config 2>/dev/null --cflags nss` `pkg-config
2>/dev/null --exists nss && echo -DHAVE_NSS` `find 2>&1 /usr/include
/usr/local/include -name "gmp.h" | grep -q "gmp.h" && echo -DHAVE_GMP`
`find 2>&1 /usr/include/ /usr/local/include/ -name "krb5.h" | grep -q
"krb5.h" && echo -DHAVE_KRB5`     -funroll-loops SIPdump.c -o SIPdump.o
gcc SIPdump.o -s -L/usr/local/lib -L/usr/local/ssl/lib -lssl -lcrypto -lm
-lz  `pkg-config 2>/dev/null --libs nss` `find 2>&1 /usr/include
/usr/local/include -name "gmp.h" | grep -q "gmp.h" && echo -lgmp` `find
2>&1 /usr/include/ /usr/local/include/ -name "krb5.h" | grep -q "krb5.h" &&
echo -lk5crypto -lkrb5`  -lpcap -o ../run/SIPdump
ld: warning: directory not found for option '-L/usr/local/ssl/lib'
ld: warning: option -s is obsolete and being ignored
gcc -Wall vncpcap2john.c -lpcap -o ../run/vncpcap2john
All done

$ ./john  --format=dmg-opencl --test=30
Device 1: GeForce GT 750M
Local worksize (LWS) 64, Global worksize (GWS) 9216
Benchmarking: dmg-opencl, Apple DMG [PBKDF2-SHA1 3DES/AES OpenCL]... DONE
Raw:    14623 c/s real, 34002 c/s virtual


$ ./john  --format=dmg-opencl  --wordlist=/WList/run/output_single.txt
/Users/kris/Dropbox/Public/jeff-sparsebundle-hash/hash
Device 1: GeForce GT 750M
Local worksize (LWS) 64, Global worksize (GWS) 9216
Loaded 1 password hash (dmg-opencl, Apple DMG [PBKDF2-SHA1 3DES/AES OpenCL])
Press 'q' or Ctrl-C to abort, almost any other key for status
Abort trap: 6

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.