Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[HIPIFY][feature] Add support of distinguishing overload functions #662

Open
emankov opened this issue Oct 20, 2022 · 0 comments
Open

[HIPIFY][feature] Add support of distinguishing overload functions #662

emankov opened this issue Oct 20, 2022 · 0 comments
Assignees
Labels
feature Feature request or implementation

Comments

@emankov
Copy link
Collaborator

emankov commented Oct 20, 2022

[Case study]

__CUDA_HOSTDEVICE_FP16_DECL__ float __low2float(const __half2 a);

is supported by HIP:

    inline float __low2float(__half2 x);

While

__CUDA_HOSTDEVICE_BF16_DECL__ float __low2float(const __nv_bfloat162 a);

is not.

[ToDo]

  • Introduce a corresponding marker(s)
  • Introduce a corresponding matcher(s)
  • Provide tests for the feature
  • [Long Term] Revise all the overridden functions (might be split into separate activities)
@emankov emankov added the feature Feature request or implementation label Oct 20, 2022
@emankov emankov self-assigned this Oct 20, 2022
emankov added a commit to emankov/HIPIFY that referenced this issue Oct 22, 2022
+ Updated the regenerated hipify-perl and CUDA_Device_API_supported_by_HIP.md accordingly

[ToDo]
+ There are plenty of overridden `bfloat16` functions which are not supported yet; so take them first into account while implementing ROCm#662
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
feature Feature request or implementation
Projects
None yet
Development

No branches or pull requests

1 participant