[Perf] Linux/arm64: 3 Improvements on 5/19/2026 11:22:03 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [051156cd1ce17060fd599e1a19f7d1bfb44519f8](https://github.com/dotnet/runtime/commit/051156cd1ce17060fd599e1a19f7d1bfb44519f8)
Compare | [c6a8e4caf07c58141038ca47aedc61ff39d33d96](https://github.com/dotnet/runtime/commit/c6a8e4caf07c58141038ca47aedc61ff39d33d96)
Diff | [Diff](https://github.com/dotnet/runtime/compare/051156cd1ce17060fd599e1a19f7d1bfb44519f8...c6a8e4caf07c58141038ca47aedc61ff39d33d96)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>

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

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

| 403.13 ns | 341.57 ns | 0.85 | 0.11 | False | | |
|

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

| 402.61 ns | 347.12 ns | 0.86 | 0.11 | False | | |
|

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

| 398.71 ns | 349.74 ns | 0.88 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives*'
```

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.Max_Vector(BufferLength: 3079)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.Add_Vector(BufferLength: 3079)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.BitwiseAnd_Vector(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 linked benchmark source, src/benchmarks/micro/libraries/System.Numerics.Tensors/Perf_NumberTensorPrimitives.cs, and review the Max_Vector, Add_Vector, and BitwiseAnd_Vector cases. Run the documented benchmarks_ci.py command and compare the ARM64 Ubuntu results with the linked baseline and test reports. Done means the reported performance change is explained and any needed follow-up is identified.

Written by the indexing model from the issue text.