Make Math/MathF.Sign inlineable for floating point data

[MichalPetryka]

Description

Math/MathF.Sign for floating point data currently isn't inlined, probably due to the throw there. Since it's a simple math operation, it should probably be inlined.

Configuration

Sharplab Core CLR 5.0.721.25508 on amd64

Regression?

No idea.

Data

Sharplab

Analysis

Moving the throw there to a throw helper should probably make it inlineable.

[dotnet-issue-labeler]

I couldn't figure out the best area label to add to this issue. If you have write-permissions please help me learn by adding exactly one area label.

[ghost]

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

Issue Details

---

Description

Math/MathF.Sign for floating point data currently isn't inlined, probably due to the throw there. Since it's a simple math operation, it should probably be inlined.

Configuration

Sharplab Core CLR 5.0.721.25508 on amd64

Regression?

No idea.

Data

Sharplab

Analysis

Moving the throw there to a throw helper should probably make it inlineable.

Author:	MichalPetryka
Assignees:	-
Labels:	area-System.Numerics, tenet-performance
Milestone:	-

[tannergooding]

It's probably not worthwhile to make this inlineable. If perf is a consideration (and even if it isn't) then double/float.IsNegative or Math/MathF.CopySign is probably better.

The only consideration really is that these correctly report or propagate -0 and -NaN (that is, IsNegative reports true for negative zero and negative NaN; and CopySign will do the same, copying the sign to or from such inputs).

[dakersnar]

Closing this for reasons explained above.