[Perf] Windows/x64: 6 Regressions on 5/2/2026 8:54:14 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [2e6efe3a93c7a685d76951a7bfd91c520dc7b333](https://github.com/dotnet/sdk/commit/2e6efe3a93c7a685d76951a7bfd91c520dc7b333)
Compare | [5d0de746ed8accbd81c2a5c3c7bfa4e70b2df80e](https://github.com/dotnet/sdk/commit/5d0de746ed8accbd81c2a5c3c7bfa4e70b2df80e)
Diff | [Diff](https://github.com/dotnet/sdk/compare/2e6efe3a93c7a685d76951a7bfd91c520dc7b333...5d0de746ed8accbd81c2a5c3c7bfa4e70b2df80e)
Configs | CompilationMode:tiered, RunKind:bepuphysics

### Regressions in DemoBenchmarks.TwoBodyConstraintBenchmarks

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

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

| 43.96 μs | 60.92 μs | 1.39 | 0.00 | True | | |
|

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

| 115.25 μs | 137.36 μs | 1.19 | 0.00 | True | | |
|

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

| 82.29 μs | 98.05 μs | 1.19 | 0.00 | True | | |
|

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

| 68.33 μs | 79.97 μs | 1.17 | 0.00 | True | | |
|

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

| 84.73 μs | 98.40 μs | 1.16 | 0.00 | True | | |
|

• [Contact4Nonconvex - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 483.85 μs | 541.28 μs | 1.12 | 0.00 | True | | |

![graph]()
![graph]()
![graph]()
![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 'DemoBenchmarks.TwoBodyConstraintBenchmarks*'
```

### DemoBenchmarks.TwoBodyConstraintBenchmarks.AngularSwivelHinge

#### ETL Files

#### Histogram

#### JIT Disasms

### DemoBenchmarks.TwoBodyConstraintBenchmarks.AngularHinge

#### ETL Files

#### Histogram

#### JIT Disasms

### DemoBenchmarks.TwoBodyConstraintBenchmarks.DistanceServo

#### ETL Files

#### Histogram

#### JIT Disasms

### DemoBenchmarks.TwoBodyConstraintBenchmarks.BallSocket

#### ETL Files

#### Histogram

#### JIT Disasms

### DemoBenchmarks.TwoBodyConstraintBenchmarks.LinearAxisServo

#### ETL Files

#### Histogram

#### JIT Disasms

### DemoBenchmarks.TwoBodyConstraintBenchmarks.Contact4Nonconvex

#### 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 at the six benchmark ranges linked in the report, then run the provided benchmarks_ci.py command. Compare the baseline and test commits and inspect the linked test report and profiling workflow docs. Done means the reported Windows/x64 regressions are explained and no longer reproduced in the benchmark results.

Written by the indexing model from the issue text.