[Perf] Windows/x86: 2 Regressions on 7/27/2026 4:51:30 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | x86
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [ca073c0866e1504313b43781c209f7f680d0ecb3](https://github.com/dotnet/runtime/commit/ca073c0866e1504313b43781c209f7f680d0ecb3)
Compare | [65b0190679e7afaea3c09bcfc474c55414403972](https://github.com/dotnet/runtime/commit/65b0190679e7afaea3c09bcfc474c55414403972)
Diff | [Diff](https://github.com/dotnet/runtime/compare/ca073c0866e1504313b43781c209f7f680d0ecb3...65b0190679e7afaea3c09bcfc474c55414403972)
Configs | CompilationMode:tiered, RunKind:micro

### Regressions in System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Single>

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

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

| 131.96 ns | 156.37 ns | 1.19 | 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 'System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives*'
```

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Single>.SumOfMagnitudes(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)

---

### Run Information

Name | Value
-- | --
Architecture | x86
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [ca073c0866e1504313b43781c209f7f680d0ecb3](https://github.com/dotnet/runtime/commit/ca073c0866e1504313b43781c209f7f680d0ecb3)
Compare | [65b0190679e7afaea3c09bcfc474c55414403972](https://github.com/dotnet/runtime/commit/65b0190679e7afaea3c09bcfc474c55414403972)
Diff | [Diff](https://github.com/dotnet/runtime/compare/ca073c0866e1504313b43781c209f7f680d0ecb3...65b0190679e7afaea3c09bcfc474c55414403972)
Configs | CompilationMode:tiered, RunKind:micro

### Regressions in System.Memory.Span<Char>

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

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

| 8.05 ns | 9.44 ns | 1.17 | 0.40 | 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 'System.Memory.Span*'
```

### System.Memory.Span<Char>.StartsWith(Size: 512)

#### 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 reported benchmark sources in src/benchmarks/micro/libraries/System.Numerics.Tensors/Perf_NumberTensorPrimitives.cs and src/benchmarks/micro/libraries/System.Memory/Span.cs. Run the provided benchmarks_ci.py commands for the two filters on the Windows x86 configuration, then compare the linked baseline and test reports. Done means the cause of both regressions is identified and the performance results no longer regress.

Written by the indexing model from the issue text.