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

Question about current state of vectorized paths for Log/Pow/Cbrt/RootN in System.Numerics.Tensors #111461

Closed
alexcovington opened this issue Jan 15, 2025 · 3 comments

Comments

@alexcovington
Copy link
Contributor

I saw a note saying the vectorized paths for Log/Pow/Cbrt/RootN in System.Numerics.Tensors were disabled because of a bug. I looked through open issues, but I couldn't find any mention of a resolution for the bugs in the vectorized path. Wondering if a resolution is planned for .NET 10.0?

@dotnet-policy-service dotnet-policy-service bot added the untriaged New issue has not been triaged by the area owner label Jan 15, 2025
Copy link
Contributor

Tagging subscribers to this area: @dotnet/area-system-numerics
See info in area-owners.md if you want to be subscribed.

@stephentoub
Copy link
Member

A robust vectorized implementation would be welcome. There are a bunch of TensorPrimitives methods that are still not vectorized, including these. #97193 tracks that.

@dotnet-policy-service dotnet-policy-service bot removed the untriaged New issue has not been triaged by the area owner label Jan 15, 2025
@tannergooding
Copy link
Member

@alexcovington Worth noting in many cases we're using https://github.com/amd/aocl-libm-ose and simply haven't ported these over yet.

If the AOCL LIBM has an existing algorithm that is the most preferred; otherwise we'd need to spend the time rolling our own algorithm.

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

No branches or pull requests

3 participants