Support C++ AMP about glm

[Groovounet]

It's actually a lot more complex than what you were expecting. C++ AMP doesn't support 16 bits integers (Used for half) or built-in std functions like "floor".

It sounds very unlikely that such port happen unless C++ AMP became used enough that the work required to support it happen.

Thanks,
Christophe

[ssylvan]

Well, you could just leave the restrict(amp) thing off of any functions that wouldn't compile with it. Note: I'm not suggesting you actually implement any algorithms using AMP, just that you make it possible for others to use GLM in their AMP algorithms. If they can't use half or floor until AMP supports it, then so be it.

[Groovounet]

if you willing to provide a pull request for GLM 0.9.5 branch, I would be very happy to accept it.

Thanks,
Christophe