[Perf] Linux/arm64: 3 Improvements on 6/22/2026 4:29:08 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [4cb447e6c483c8845a2e91c6c319c7424216b057](https://github.com/dotnet/runtime/commit/4cb447e6c483c8845a2e91c6c319c7424216b057)
Compare | [59979e6401af21c7cdf4f61fb35ea3568cb93e5d](https://github.com/dotnet/runtime/commit/59979e6401af21c7cdf4f61fb35ea3568cb93e5d)
Diff | [Diff](https://github.com/dotnet/runtime/compare/4cb447e6c483c8845a2e91c6c319c7424216b057...59979e6401af21c7cdf4f61fb35ea3568cb93e5d)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Improvements in System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives<Double>

Benchmark | Baseline | Test | Test/Base | Test Quality | Edge Detector | Baseline IR | Compare IR | IR Ratio
-- | -- | -- | -- | -- | -- | -- | -- | --
|

• [Sinh - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 151.60 μs | 139.08 μs | 0.92 | 0.18 | False | | |
|

• [CosineSimilarity - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 777.92 μs | 715.86 μs | 0.92 | 0.11 | False | | |
|

• [Sin - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 2.88 ms | 2.68 ms | 0.93 | 0.13 | False | | |

![graph]()
![graph]()
![graph]()
[Test Report]()

### Repro
General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md

```cmd
git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives*'
```

### System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives<Double>.Sinh(BufferLength: 128)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives<Double>.CosineSimilarity(BufferLength: 3079)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives<Double>.Sin(BufferLength: 3079)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

[Profiling workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/profiling-workflow-dotnet-runtime.md)
[Benchmarking workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/benchmarking-workflow-dotnet-runtime.md)

Start with the benchmark source at performance/src/benchmarks/micro/libraries/System.Numerics.Tensors/Perf_FloatingPointTensorPrimitives.cs and the linked report for Sinh, CosineSimilarity, and Sin. Run the provided benchmarks and compare the listed baseline and test results on Linux/arm64. The report does not define a code change or a clear completion condition, so further triage is needed before implementation.

Written by the indexing model from the issue text.