[Perf] Linux/arm64: 3 Regressions on 7/8/2026 8:49:11 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [c76054f5cb1aae6ab146ec7a8db70ca3aeda4ac9](https://github.com/dotnet/runtime/commit/c76054f5cb1aae6ab146ec7a8db70ca3aeda4ac9)
Compare | [3ade0b55c797a31f6800040e3ae3c5c75bc0d2dc](https://github.com/dotnet/runtime/commit/3ade0b55c797a31f6800040e3ae3c5c75bc0d2dc)
Diff | [Diff](https://github.com/dotnet/runtime/compare/c76054f5cb1aae6ab146ec7a8db70ca3aeda4ac9...3ade0b55c797a31f6800040e3ae3c5c75bc0d2dc)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in System.Numerics.Tests.Perf_Matrix3x2

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

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

| 9.75 ns | 975.10 ns | 100.00 | 0.08 | False | | |
|

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

| 9.78 ns | 973.87 ns | 99.62 | 0.12 | False | | |
|

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

| 2.66 ns | 11.97 ns | 4.50 | 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.Tests.Perf_Matrix3x2*'
```

### System.Numerics.Tests.Perf_Matrix3x2.MultiplyByMatrixBenchmark

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tests.Perf_Matrix3x2.MultiplyByMatrixOperatorBenchmark

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tests.Perf_Matrix3x2.InvertBenchmark

#### 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 src/benchmarks/micro/libraries/System.Numerics.Vectors/Perf_Matrix3x2.cs and the benchmarks_ci.py command in the Repro section. Compare the listed baseline and test commits, then inspect the linked test report and ARM64 benchmark results. Done means the cause of the three regressions is identified and the affected measurements are no longer regressed.

Written by the indexing model from the issue text.