Yunsong Wang
Yunsong Wang
cuCo doesn't have a release yet thus I'm not sure how to incorporate the versioned inline namespace in our case. e.g. how to name the version.
@jrhemstad is working on a base `set` class and using `atomic_ref` is part of this exploration IIRC. By all means, NVIDIA/libcudacxx#98 is already bulky enough thus a separate PR to...
> What's the state of this PR? This could be used to automatically catch bugs such as #510. I vaguely recall that some upstream libraries couldn't be built with clang....
@andrewbriand Can you resolve the conflict in CMake?
ok to test
@andrewbriand Can you please also merge with the latest `dev` branch and fix build warnings (if there is any)?
Another improvement we should try: using one CG to insert/retrieve multiple keys instead of a single key to amortize CG overhead.
One reduce instead of two ballots to potentially improve `contains`: https://godbolt.org/z/qv6Ghv8M3 Not necessarily improve the performance, worth testing though
Updates: Building cuCollections in WSL is pretty straightforward. Once CUDA and a relatively new CMake are installed, all examples, benchmarks and unit tests can build and run flawlessly. Note: the...