NVIDIA / NVIDIA/cccl

`cuda::annotated_ptr` header can't be included in host code

Open
#1,019 3 comments 0 reactions 0 assignees View on GitHub
Dominant language
C++
Stars
2.5k
Forks
486
Avg merge
2d 6h
Merged PRs (30d)
295

Description

When including the header cuda/annotated_ptr in host code the following error is shown:

```
/usr/local/cuda/targets/x86_64-linux/include/cuda/std/barrier: In static member function 'static cuda::__4::async_contract_fulfillment cuda::__4::__memcpy_async_impl<4, false>::__copy(char*, const char*, size_t, size_t, size_t)':
/usr/local/cuda/targets/x86_64-linux/include/cuda/std/barrier:614:51: error: '__cvta_generic_to_shared' was not declared in this scope
614 | :: "r"(static_cast(__cvta_generic_to_shared(__destination + __offset))),
```

Among others. It would be useful to support this since users can define structures with this type in it that's used on both the host and device.

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.