Using AMD APP SDK with C#? [closed]

Question

I have searched everywhere but couldn't find any definite information.
Are there any bindings or a support for using AMD APP SDK with C#?
I have seen that there is APARAPI for Java, does something similar exist for C#?

Main reason why I'm asking this is that I would like to try out HSA and hUMA features, but I don't want to waste time with C++.

Answer 1

Take a look at this option GpuLinq It compiles LINQ code from C# or F# source to OpenCL. This isn't a generic C# API, but executes LINQ in parallel.,

From their Github Web page

"GpuLinq's main mission is to democratize GPGPU programming through LINQ. The main idea is that we represent the query as an Expression tree and after various transformations-optimizations we compile it into fast OpenCL kernel code. In addition we provide a very easy to work API without the need of messing with the details of the OpenCL API."

F# is a better fit for parallel programming on GPU's than C#, because is has parallel asynchronous patterns.

As of October 2014, HSA and hUMA are still very cutting edge. The Linux Kernel patch was in June 2014, Java support is coming in 2015, There may be problems with a C#/F# to OpenCL layered approach that are untested and untried with HSA. So trying to use a higher-level programming language may lead you to spend more time debugging infrastructure issues, than you would spend coding in C or C++.

Answer 2

I found this: OpenCL.net .NET bindings for OpenCL that are easy-to-use and true to the original API. There is no OOP abstraction, nor will there ever be. OpenCL.Net is meant to be small, fast (with as little explicit marshaling as possible) and .NET friendly at the same time.

https://www.nuget.org/packages/OpenCL.Net/