[Perf] Windows/x86: 1 Regression on 2/22/2026 9:59:12 PM +00:00

### Run Information

Name | Value
-- | --
Architecture | x86
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [5302011c5d68c1b0b98fa8599d0a14cad979344b](https://github.com/dotnet/runtime/commit/5302011c5d68c1b0b98fa8599d0a14cad979344b)
Compare | [0ff5be7a1f7f560c1f3b7454fa9d4c5ec03512a8](https://github.com/dotnet/runtime/commit/0ff5be7a1f7f560c1f3b7454fa9d4c5ec03512a8)
Diff | [Diff](https://github.com/dotnet/runtime/compare/5302011c5d68c1b0b98fa8599d0a14cad979344b...0ff5be7a1f7f560c1f3b7454fa9d4c5ec03512a8)
Configs | CompilationMode:tiered, RunKind:micro

### Regressions in PerfLabTests.CastingPerf2.CastingPerf

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

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

| 135.46 μs | 151.73 μs | 1.12 | 0.29 | False | | |

[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 'PerfLabTests.CastingPerf2.CastingPerf*'
```

### PerfLabTests.CastingPerf2.CastingPerf.FooObjIsDescendant

#### 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 source in src/benchmarks/micro/runtime/perflab/CastingPerf2.cs, especially lines 131-136, and read the benchmarking workflow linked in the issue. Reproduce PerfLabTests.CastingPerf2.CastingPerf with the provided benchmarks_ci.py command, then compare the baseline and test commits and inspect the linked history and report. Done means the x86 Windows regression is explained and addressed or shown to no longer reproduce.

Written by the indexing model from the issue text.