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

Extend cuda::std::numeric_limits for __half, __nv_bfloat16, __nv_fp8_e4m3, __nv_fp8_e5m2 and deprecate cub::FpLimits #3349

Open
Tracked by #101
bernhardmgruber opened this issue Jan 10, 2025 · 3 comments

Comments

@bernhardmgruber
Copy link
Contributor

bernhardmgruber commented Jan 10, 2025

Part of this is covered here: #525

@fbusato
Copy link
Contributor

fbusato commented Jan 13, 2025

would make sense to also add the following functions for completeness?

  • isfinite
  • isinf
  • isnan
  • isnormal
  • signbit

(@davebayer for visibility)

@miscco
Copy link
Collaborator

miscco commented Jan 14, 2025

Can we split __nv_fp8_e4m3 and __nv_fp8_e5m2 into their separate issue because we currently do not support them at all in libcu++

@bernhardmgruber
Copy link
Contributor Author

would make sense to also add the following functions for completeness?

  • isfinite
  • isinf
  • isnan
  • isnormal
  • signbit

Eventually yes, but I don't need them to deprecate/replace cub::Traits, so they are less urgent IMO.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
Status: Todo
Development

No branches or pull requests

3 participants