Merge pull request #1250 from leofang/cuda_compiler

Add FAQ to explain CF's CUDA compiler is not intended to be used by end users

Author: beckermr

src/user/faq.rst

@@ -71,4 +71,10 @@ FAQ
 
     x86_64-apple-darwin13.4.0-clang: No such file or directory
     
-  are a telltale sign that you are lacking compilers.
+  are a telltale sign that you are lacking compilers.
+
+:ref:`(Q) <faq_cuda_compiler_header>` **How can I compile CUDA (host or device) codes in my environment?**
+
+  Unfortunately, this is not possible with Conda-Forge's current infrastructure (``nvcc``, ``cudatoolkit``, etc) if there is no local CUDA Toolkit installation. In particular, the ``nvcc`` package provided on Conda-Forge is a *wrapper package* that exposes the actual ``nvcc`` compiler to our CI infrastructure in a ``conda``-friendly way; it does not contain the full ``nvcc`` compiler toolchain. One of the reasons is that CUDA headers like ``cuda.h``, ``cuda_runtime.h``, etc, are not redistributable according to NVIDIA's EULA, which are needed at compile time. Likewise, the ``cudatoolkit`` package only contains CUDA runtime libraries and not the compiler toolchain.
+  
+  If you need to compile CUDA codes, even if they involve only CUDA host APIs, you will still need a valid CUDA Toolkit installed locally and use it. Please refer to `NVCC's documentation <https://docs.nvidia.com/cuda/cuda-compiler-driver-nvcc/index.html>`_ for the CUDA compiler usage and `CUDA Programming Guide <https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html>`_ for general CUDA programming.