NVIDIA / NVIDIA/cccl

Settle on an API for `cuda::vector`

Open
#2,058 0 comments 0 reactions 1 assignee Claimed by @miscco View on GitHub
Dominant language
C++
Stars
2.5k
Forks
486
Avg merge
2d 6h
Merged PRs (30d)
295

Description

We want to introduce `cuda::vector` so that it can serve as a drop-replacement for most user code.

Therefore it is essential, that we provide the facilities that users expect from a standard container. Recently `std::inplace_vector` was added to C++26. That is a special container in that it does not take an allocator similar to what we envision with `cuda::vector`.

We should make a decision whether we want to deviate from the API of `std::inplace_vector` or whether we just want to take it as is.

The advantage is that it is highly familiar to users and has already gone through standardization so the design is sound.
The disadvantage is that it does carry some cruft from all the standard containers with it.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.