Cycles:

* Add back option to bundle CUDA kernel binaries with builds.
* Disable runtime CUDA kernel compilation on Windows, couldn't get this working,
  since it seems to depend on visual studio being installed, even though for
  this particular case it shouldn't be needed. CMake only at the moment.
* Runtime compilation on linux/mac should now work if nvcc is not installed in
  the default location, but available in PATH.

intern/cycles/device/device_cuda.cpp

@@ -223,6 +223,10 @@ public:
 		if(path_exists(cubin))
 			return cubin;
 
+#ifdef WITH_CUDA_BINARIES
+		fprintf(stderr, "CUDA binary kernel for this graphics card not found.\n");
+		return "";
+#else
 		/* if not, find CUDA compiler */
 		string nvcc = cuCompilerPath();
 
@@ -260,6 +264,7 @@ public:
 		printf("Kernel compilation finished in %.2lfs.\n", time_dt() - starttime);
 
 		return cubin;
+#endif
 	}
 
 	bool load_kernels()