Hi Levent,

...

The whole point of the Xeon-phi is the availability of large-vector sized floating-point support and many-many cores.

Sure, that's what I'm contemplating - use the many options of writing parallel and concurrent Haskell programs.

So, GHC's RTS should "just work"?

I was hoping someone already had actually seen this on their machine.

- J.

I recall we thought about using a Xeon phi for our application but abandoned the attempt for reasons I cannot recall. I’ll see if I can find out. But maybe you would better using something like accelerate-llvm? I think you wrote your application was numerical. BTW I am currently adding a small improvement to ghc floating point code generation but I am beginning to think it’s going to be difficult to get ghc to generate really good numerical / floating point code and I am experimenting with accelerate-llvm. Dominic