![]() ![]() Step 1: Install the CUDA driver and toolkit (You can install them if you want but I have not tested it with these instructions.) Boost is not required at all so you don't need the Boost C++ libraries or the Boost.Python libraries.(I used the i686-apple-darwin11-llvm-gcc-4.2 compiler that came with Xcode) A C++ compiler, Installing Xcode will take care of this.Optional - Nvidia's GPU Computing SDK here (Has some nice samples in C++ etc.).here (CUDA 4.0 is current as of this entry) Nvidia's CUDA Developer Drivers here (Version 4.0.50 is current as of this entry).This is because many of the earlier P圜UDA dependencies now come preinstalled with 10.7 Below are the components used to build/use P圜UDA: Installing P圜UDA on OS X 10.7 is a bit easier than with 10.6 and other versions. ( BogdanVacaliuc) Installing P圜UDA on Mac OS X 10.7 Lion with CUDA 4.0 See also: py.test 2.0.0 PyOpenCL Installation for Mac $ cd pycuda-VERSION/test # if you're not there already Simply install pytest before running the confidence tests. You might have gotten this if you install P圜UDA before PyOpenCL. % install_name_tool -change /System/Library/Frameworks/amework/Versions/2.6/Python /Library/Frameworks/amework/Versions/2.6/Python libboost_python.dylib For example, the following command will relink from the System Python (2.6.1) to the MacPython (2.6.4) on Snow Leopard: ![]() If for some reason you are unable to get Boost to link against the correct python, you can use install_name_tool to relink it against the correct one. ( BryanCatanzaro) Fatal Python error: Interpreter not initialized (version mismatch?) Or you can make this change permanent and global by writing this to the defaults:ĭefaults write Prefer-32-Bit 1 If you'd like to, you can force Python on Snow Leopard to run in 32-bit by setting a special environment variable: If it's a huge number, you've got a 64-bit Python. If it's ~2 billion, you're running Python in 32-bit mode. Python on Snow Leopard defaults to a 64-bit executable. If it says "OK" at the end, that means your installation of P圜UDA was successful. If you'd like to be extra-careful, you can run P圜UDA's unit tests: These packages are listed under install_requires in P圜UDA's install control file. ![]() Note that this step may automatically look for and, if necessary, install a number of Python packages from the Python package index. Once that works, congratulations! You've successfully built P圜UDA. LDFLAGS.extend()Įnsure that there are no warnings around the line of "different architecture" # if on Snow Leopard, include these lines:ĬXXFLAGS = $ python configure.py -cuda-root=/usr/local/cuda/ $ cd pycuda-VERSION # if you're not there already Don't repeat these commands for the Mac system Python. If you're not sure try import numpy at the Python prompt. $ sudo "easy_install numpy" # this will install numpy using distribute $ sudo "python distribute_setup.py" # this will install distribute ![]() Here's an easy way to install it, if you do not have it already: P圜UDA is designed to work in conjunction with numpy, Python's array package. Read below why we want to run python in 32-bit, but basically it has to do with the fact that CUDA 2.2 currently only supports 32-bit mode, and hence P圜UDA as well. It may be in your best interest to stick with this version for now until someone else can show us how to run a newly installed version of MacPython in 32-bit mode. Snow Leopard comes installed with Python 2.6.1.However, Xcode will take 2+ GB of your drive.) (You can install Xcode and not bother installing the C++ compiler and Mac OS SDK separately. (Python 2.5.1 comes pre-installed on Mac OS X should work just fine.) MacOSX10.4u.sdk from Apple Developer Tools package.A C++ compiler, preferably a Version 4.x gcc.CUDA 2.2 is the latest one to date and it works fine in OS X.a Mac OS 10.5.7 (previous versions should work fine.).Below are the components used to build/use P圜UDA: In fact, you can use the shell (Terminal) for all operations. The process of installing P圜UDA on Mac OS X is very similar to the Linux one. If you're running Snow Leopard (Mac OS 10.6), then you might benefit from following these slightly more up-to-date instructions: ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |