cuda-api-wrappers
cuda-api-wrappers copied to clipboard
Add methods for getting individual kernel attributes - to the runtime-API-only branch
The Driver-API-based branch already obtains kernel properties individually. Let's have such methods in kernel_t for the runtime API, so that the user does not have to know about what kernel::attribute_t even is, and can just use methods of kernel_t. Now, this will be less efficient with the runtime API if we repeatedly get single attributes, but if that's a problem one can still get all of the attributes.