Runtime Error with a HW configuration 8600GT 128MB

Hi Folks

I am excited to have CUDA in my favorite platform.

My HW configuration
MacBook Pro
Chipset Model: GeForce 8600M GT
Type: Display
Bus: PCIe
PCIe Lane Width: x16
VRAM (Total): 128 MB

My CUDA setup :
Toolkit at /usr/local/cuda
And SDK at /usr/local/cuda/sdk

I bump into the following error. I am new to Macs so sorry if it’s basic. In the commandline
Prompt>./simpleGL
dyld: Library not loaded: @rpath/libcuda.dylib
Referenced from: /usr/local/cuda/sdk/bin/macosx/release/./simpleGL
Reason: image not found
Trace/BPT trap

Even if I “make” simpleGL ( or any other project ), the error is the same.
Thanks.

You need to set the DYLD_LIBRARY_PATH.

For example , add these two lines to your .bash_profile file:

export PATH=/usr/local/cuda/bin:$PATH
export DYLD_LIBRARY_PATH=/usr/local/cuda/lib:$DYLD_LIBRARY_PATH

Exit the terminal or reload the .bash_profile ( . .bash_profile).

You only have 128MB of VRAM, some of the examples will fail.

Thanks much. It worked. I have written a project which is flexible enough to scale according to the hardware configuration. Basically developing in this 128 MB and deploying it in a much powerful CUDA setup.