[Perf] Linux/arm64: 4 Improvements on 6/30/2026 7:21:17 PM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [dc6840e6978b3f0e3f2917b43f4b27a05f3cc984](https://github.com/dotnet/runtime/commit/dc6840e6978b3f0e3f2917b43f4b27a05f3cc984)
Compare | [01890b5bf452cfe1fea89cf3302b93a2950117ca](https://github.com/dotnet/runtime/commit/01890b5bf452cfe1fea89cf3302b93a2950117ca)
Diff | [Diff](https://github.com/dotnet/runtime/compare/dc6840e6978b3f0e3f2917b43f4b27a05f3cc984...01890b5bf452cfe1fea89cf3302b93a2950117ca)
Configs | CompilationMode:tiered, RunKind:sve

### Improvements in SveBenchmarks.PairwiseAdd

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

• [Sve2PairwiseAdd - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 289.14 ns | 178.04 ns | 0.62 | 0.23 | False | | |
|

• [Sve2PairwiseAdd - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 5.40 μs | 4.43 μs | 0.82 | 0.21 | False | | |
|

• [Sve2PairwiseAdd - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 74.45 ns | 62.02 ns | 0.83 | 0.25 | 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 'SveBenchmarks.PairwiseAdd*'
```

### SveBenchmarks.PairwiseAdd.Sve2PairwiseAdd(Size: 527)

#### ETL Files

#### Histogram

#### JIT Disasms

### SveBenchmarks.PairwiseAdd.Sve2PairwiseAdd(Size: 10015)

#### ETL Files

#### Histogram

#### JIT Disasms

### SveBenchmarks.PairwiseAdd.Sve2PairwiseAdd(Size: 127)

#### 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)

---

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [dc6840e6978b3f0e3f2917b43f4b27a05f3cc984](https://github.com/dotnet/runtime/commit/dc6840e6978b3f0e3f2917b43f4b27a05f3cc984)
Compare | [01890b5bf452cfe1fea89cf3302b93a2950117ca](https://github.com/dotnet/runtime/commit/01890b5bf452cfe1fea89cf3302b93a2950117ca)
Diff | [Diff](https://github.com/dotnet/runtime/compare/dc6840e6978b3f0e3f2917b43f4b27a05f3cc984...01890b5bf452cfe1fea89cf3302b93a2950117ca)
Configs | CompilationMode:tiered, RunKind:sve

### Improvements in SveBenchmarks.MultiplyPow2

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

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

| 89.98 ns | 52.33 ns | 0.58 | 0.20 | False | | |

![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 'SveBenchmarks.MultiplyPow2*'
```

### SveBenchmarks.MultiplyPow2.SveTail(Size: 127)

#### 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 sources src/benchmarks/micro/sve/PairwiseAdd.cs and MultiplyPow2.cs, then read the benchmarking and profiling workflow documents linked in the report. Run benchmarks_ci.py with the listed filters and compare the ARM64 SVE results against the provided baseline and test reports. Done means identifying and documenting the cause or disposition of the reported performance changes.

Written by the indexing model from the issue text.