Hacker News Reader

It's basically defunctionalization. It's hard to get access to function pointers on the GPU to do "real" higher order functions.

What version of CUDA are you talking about. I though the latest version supported indirect function pointers (finally).

That's good news! Did that change with Kepler? Any links to more info?

I wish they'd at least give us a computed goto...but alas, every switch is destined for predicated branching.