[Perf] Windows/x64: 3 Regressions on 8/30/2026 9:30:50 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [100002779ceec649fcfd00c233f673b494130827](https://github.com/dotnet/sdk/commit/100002779ceec649fcfd00c233f673b494130827)
Compare | [17974d44ef211135d3e9b0b84f5e03830dd59ad8](https://github.com/dotnet/sdk/commit/17974d44ef211135d3e9b0b84f5e03830dd59ad8)
Diff | [Diff](https://github.com/dotnet/sdk/compare/100002779ceec649fcfd00c233f673b494130827...17974d44ef211135d3e9b0b84f5e03830dd59ad8)
Configs | CompilationMode:tiered, RunKind:bepuphysics

### Regressions in DemoBenchmarks.TwoBodyConstraintBenchmarks

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

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

| 168.07 μs | 203.01 μs | 1.21 | 0.00 | True | | |
|

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

| 49.37 μs | 57.43 μs | 1.16 | 0.00 | True | | |

![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.Weld

#### ETL Files

#### Histogram

#### JIT Disasms

### DemoBenchmarks.TwoBodyConstraintBenchmarks.SwingLimit

#### 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 | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [100002779ceec649fcfd00c233f673b494130827](https://github.com/dotnet/sdk/commit/100002779ceec649fcfd00c233f673b494130827)
Compare | [17974d44ef211135d3e9b0b84f5e03830dd59ad8](https://github.com/dotnet/sdk/commit/17974d44ef211135d3e9b0b84f5e03830dd59ad8)
Diff | [Diff](https://github.com/dotnet/sdk/compare/100002779ceec649fcfd00c233f673b494130827...17974d44ef211135d3e9b0b84f5e03830dd59ad8)
Configs | CompilationMode:tiered, RunKind:bepuphysics

### Regressions in DemoBenchmarks.GroupedCollisionTesterBenchmarks

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

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

| 4.32 ms | 5.51 ms | 1.28 | 0.00 | True | | |

![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.ExpensiveCollisionBenchmarks

#### 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 benchmark sources in src/benchmarks/real-world/bepuphysics2/TwoBodyConstraintBenchmarks.cs and GroupedCollisionTesterBenchmarks.cs, then run the provided benchmarks_ci.py commands. Compare the listed baseline and test results and use the linked reports and profiling workflow docs to investigate the Windows/x64 regressions. Done means the regression is explained and addressed or clearly handed off with supporting measurements.

Written by the indexing model from the issue text.