[Perf] Windows/x64: 1 Regression on 8/8/2026 9:54:11 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [aedf05ae4b6c7f74f789e64a0a8d9d169414f04f](https://github.com/dotnet/sdk/commit/aedf05ae4b6c7f74f789e64a0a8d9d169414f04f)
Compare | [10d231dd86555cc17a757c800d821ad8e0f459d5](https://github.com/dotnet/sdk/commit/10d231dd86555cc17a757c800d821ad8e0f459d5)
Diff | [Diff](https://github.com/dotnet/sdk/compare/aedf05ae4b6c7f74f789e64a0a8d9d169414f04f...10d231dd86555cc17a757c800d821ad8e0f459d5)
Configs | CompilationMode:tiered, RunKind:bepuphysics

### Regressions in DemoBenchmarks.GroupedCollisionTesterBenchmarks

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

• [CheapCollisionBenchmarks - 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.GroupedCollisionTesterBenchmarks*'
```

### DemoBenchmarks.GroupedCollisionTesterBenchmarks.CheapCollisionBenchmarks

#### 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/GroupedCollisionTesterBenchmarks.cs and the provided benchmarks_ci.py command to reproduce CheapCollisionBenchmarks. Compare the baseline and compare commits and inspect the linked test report for the source of the Windows/x64 slowdown. Done means the regression is explained and addressed or the benchmark result is shown to be spurious.

Written by the indexing model from the issue text.