Using ccache with CUDA and nvcc

Hello,

I am developing a C++/CUDA/Python code using SWIG. I've been able to make
use of ccache within my distutils build system, but unfortunately have not
found a way to do so for CUDA using the nvcc compiler. Compiling my source
using "ccache nvcc ..." works just fine, but it doesn't appear to cache the
output.

Any help on how to use ccache for CUDA would be much appreciated!

Thanks,

Will

-- 
Will Boyd
Nuclear Science & Engineering
Massachusetts Institute of Technology
Email: [email protected]
Cell: 423.413.8469