cuda.cccl.parallel: Follow-up items for single-phase API implementation
Open
- Dominant language
- C++
- Stars
- 2.5k
- Forks
- 486
- Avg merge
- 2d 6h
- Merged PRs (30d)
- 295
Description
A few follow-up items after #5207 is merged:
- [ ] #5638
- [ ] After https://github.com/NVIDIA/cuda-python/issues/771 is resolved, switch to using `DeviceMemoryResource` again
- [ ] Allow the user to pass a custom allocator that is prioritized over the array namespace or our custom allocator
Contributor guide
Assessment
This issue has not been assessed yet.