Hi! Can I link statically with CUDA on Windows? Note that I’m not talking about building my static library, I mean namely static CUDA itself. Thanks!
yes, for the cuda runtime library (libcudart). That is the default linkage that nvcc uses
For other libraries, there may be differences
what about cuDNN … cuda by itself works … but cuDNN still unresolved and I don’t see a static lib anywhre