Skip to content

Conversation

@stephentoub
Copy link
Member

@stephentoub stephentoub commented Sep 16, 2024

The better, vectorized fix is more complex and can be done for .NET 10.

Fixes #107838

Easiest to review with whitespace hidden: https://github.com/dotnet/runtime/pull/107886/files?diff=split&w=1

The better, vectorized fix is more complex and can be done for .NET 10.
Copy link
Member

@jeffhandley jeffhandley left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thank you!

Copy link
Member

@jeffhandley jeffhandley left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

After hitting Approve, I noticed there's a failing test for the special values.

@stephentoub
Copy link
Member Author

/backport to release/9.0-rc2

@github-actions
Copy link
Contributor

Started backporting to release/9.0-rc2: https://github.com/dotnet/runtime/actions/runs/10909076272

@stephentoub stephentoub merged commit 24e7d1b into dotnet:main Sep 18, 2024
@stephentoub stephentoub deleted the fixexpvect branch September 18, 2024 02:56
sirntar pushed a commit to sirntar/runtime that referenced this pull request Sep 30, 2024
…net#107886)

* Replace VectorXx.Exp's edge case fallback with scalar processing

The better, vectorized fix is more complex and can be done for .NET 10.

* Revert addition to Helpers.IsEqualWithTolerance
@BarionLP
Copy link
Contributor

The better, vectorized fix is more complex and can be done for .NET 10.

do you guys remember this or should i create an issue for it?

@github-actions github-actions bot locked and limited conversation to collaborators Nov 16, 2024
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

TensorPrimitives.Exp<double> returns negative infinity instead of 0 for doubles between approximately -709.5 and -709.7

3 participants