[Perf] Windows/x64: 1 Regression on 8/4/2026 1:30:50 PM +00:00

### Run Information

Name | Value
-- | --
Architecture | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [813f634ceb016018f5acc9bd3c2b16e17dff4686](https://github.com/dotnet/sdk/commit/813f634ceb016018f5acc9bd3c2b16e17dff4686)
Compare | [5da2de22f7c6e2ebe7ed4c16c614b00183145d3c](https://github.com/dotnet/sdk/commit/5da2de22f7c6e2ebe7ed4c16c614b00183145d3c)
Diff | [Diff](https://github.com/dotnet/sdk/compare/813f634ceb016018f5acc9bd3c2b16e17dff4686...5da2de22f7c6e2ebe7ed4c16c614b00183145d3c)
Configs | CompilationMode:tiered, RunKind:bepuphysics

### Regressions in DemoBenchmarks.TwoBodyConstraintBenchmarks

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

• [BallSocket - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config 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 'DemoBenchmarks.TwoBodyConstraintBenchmarks*'
```

### DemoBenchmarks.TwoBodyConstraintBenchmarks.BallSocket

#### 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/real-world/bepuphysics2/TwoBodyConstraintBenchmarks.cs, especially the BallSocket benchmark, and reproduce it with the provided benchmarks_ci.py command. Compare the baseline and test commits and use the linked test report or profiling workflow to investigate the regression. Done means the BallSocket duration returns to the expected performance without regressing the benchmark.

Written by the indexing model from the issue text.