In Windows 8.1 64bit and Mathematica 10.0.2, I used the latest cuda paclet 10.0.1. However,following the sample code in CUDALink User Guide, the CUDAFunctionLoad returns an error. I can not figure what is not right. Any help?
Needs["CUDALink`"]
secondKernelCode = "
__device__ float f(float x) {
return tanf(x);
}
__global__ void secondKernel(float * diff, float h, mint listSize) {
int index = threadIdx.x + blockIdx.x * blockDim.x;
float f_n = f(((float) index) / h);
float f_n1 = f((index + 1.0) / h);
if( index < listSize) {
diff[index] = (f_n1 - f_n) / h;
}
}";
secondKernel =
CUDAFunctionLoad[secondKernelCode,
"secondKernel", {{"Float"}, "Float", _Integer}, 16]
CUDAFunctionLoad::target: "Target system specification \"TargetSystemID\" -> \!\(\"Windows-x86-64\"\) is not available for \!\(\"NVIDIA CUDA Compiler\"\) installation \!\(\"C:\\\\Users\\\\BMP\\\\AppData\\\\Roaming\\\\Mathematica\\\\Paclets\\\\Repository\\\\CUDAResources-Win64-10.0.0.1\\\\CUDAToolkit\\\\bin\\\\\"\). "