Improve `_create_void_ptr_wrapper()`
Open
cuda.compute
- Dominant language
- C++
- Stars
- 2.5k
- Forks
- 486
- Avg merge
- 2d 6h
- Merged PRs (30d)
- 295
Description
> The thinking is to use something similar to `modify_func()` below inside `_create_void_ptr_wrapper()`. My hope is that since the intrinsic `impl` is just a wrapper returning the same `codegen` function, we can create a dummy wrapper and modify its `__code__` attributes dynamically.
_Originally posted by @leofang in https://github.com/NVIDIA/cccl/pull/4249#discussion_r2066895904_
Contributor guide
Assessment
This issue has not been assessed yet.