Building CUDA kernels in parallel in VS2013?

Are there any tips for enabling parallel builds of CUDA projects in VS2013?

I’ve enabled the standard VS switches and flags but NVCC is being invoked sequentially.

I think Incredibuild is the only way to do it directly in VS2013. See also Mark Harris’s answer here

Incredibuild has a free version which limits the degree of parallelism to 8, which should be enough in most cases. I have been using it without any problem.

The only thing to note is that you need to circumvent .pdb locks with /FS specified in CUDA host compiler options:

Thanks @JanetYellen!

<1 hour later>

This works very well with CUDA. All 8 threads of a Core i7-4790 are now sweating!