What we have to do is get off the CudaToolkit on the nvidia site.
http://developer.nvidia.com/object/cuda_3_2_downloads.html
Once downloaded, run the binary as root.
$ ./cudatoolkit_3.2.9_linux_64_ubuntu10.04.run
| This places us among other things Cuda compiler. Now we must add a new route to these compilers. We can do it in ~ /. Bashrc: |
CUDA # export PATH = / usr / local / cuda / bin /: $ PATH export LD_LIBRARY_PATH =
/ usr/local/cuda/lib64 /: / usr / local / cuda / lib / Before the error
:
- $
- NVCC ejemplo.cu gcc-o example: Error Trying to exec 'cc1plus': execvp: No such file or directory
We can also download GPU Computing SDK code samples, which contains many sample programs and simulations. Once downloaded and installed, will probably have to install some more libraries. For example, to compile fluidsGL, we have to go
| / usr / bin / ld: can not find-LGL | / usr / bin / ld: can not find-LXi
make: *** [. . / .. / bin / linux / release / fluidsGL] Error 1
| $ sudo apt-get install libxmu libXi-dev-dev-dev libxt freeglut3-dev libx11-dev |
| http://developer.nvidia.com/object/cuda_3_2_downloads.html |
0 comments:
Post a Comment