C++11 wishlist for CUDA 6

This came up during a discussion at SC13, but here’s a wish list from those of us developing in the HPC unit at Penn State. Thoughts/additions welcome!

  1. std::result_of ie., compiler determination of return type for function objects
  2. range based for loop ( frowned upon by some)
  3. std::unordered_map ie., unordered associative containers (this might be available already, I should check again)
  4. lambdas

I encourage all CUDA users to file enhancements requests for the specific C++11 features that are most important to them. That way these requests can be collated, tracked, and prioritized.

Enhancement requests can be filed using the bug reporting form linked from the registered developer website. Please prefix the synopsis with “RFE:” to make enhancement requests easily distinguishable from true bugs. Thank you for your input.