Skip to content

[ci-fix] Needs review: Fix TensorPrimitives Tan precision tolerance on ARM64 iOS simulator (refs #129045) #129652

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.

Sign up for GitHub

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

Draft
github-actions wants to merge 2 commits into
base: main
Choose a base branch
from
+4 −2
Draft

[ci-fix] Needs review: Fix TensorPrimitives Tan precision tolerance on ARM64 iOS simulator (refs #129045) #129652

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f078a0c
Use explicit trig tolerance when FMA is supported to fix Tan precisio…
github-actions[bot] Jun 20, 2026
1de761f
ci: trigger checks
github-actions[bot] Jun 20, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 4 additions & 2 deletions src/libraries/System.Numerics.Tensors/tests/TensorPrimitives.Generic.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -455,7 +455,10 @@ protected override void SetSpecialValues(Span<T> x, Span<T> y)
public static IEnumerable<object[]> SpanDestinationFunctionsToTest()
{
// The current trigonometric algorithm depends on hardware FMA support for best precision.
T? trigTolerance = IsFmaSupported ? null : Helpers.DetermineTolerance<T>(doubleTolerance: 1e-10, floatTolerance: 1e-4f);
// Even with FMA, ARM64 vectorized trig can diverge a few tens of ULPs from scalar.
T? trigTolerance = IsFmaSupported
? Helpers.DetermineTolerance<T>(doubleTolerance: 1e-14, floatTolerance: 1e-5f)
: Helpers.DetermineTolerance<T>(doubleTolerance: 1e-10, floatTolerance: 1e-4f);

yield return Create(TensorPrimitives.Acosh, T.Acosh);
yield return Create(TensorPrimitives.AcosPi, T.AcosPi);
Expand Down Expand Up @@ -571,7 +574,6 @@ public void SpanDestinationFunctions_SpecialValues(SpanDestinationDelegate tenso

[Theory]
[MemberData(nameof(SpanDestinationFunctionsToTest))]
[ActiveIssue("https://github.com/dotnet/runtime/issues/124344", typeof(PlatformDetection), nameof(PlatformDetection.IsAppleMobile), nameof(PlatformDetection.IsCoreCLR))]
public void SpanDestinationFunctions_ValueRange(SpanDestinationDelegate tensorPrimitivesMethod, Func<T, T> expectedMethod, T? tolerance = null)
{
Assert.All(VectorLengthAndIteratedRange(ConvertFromSingle(-100f), ConvertFromSingle(100f), ConvertFromSingle(3f)), arg =>
Expand Down
Loading