[Perf] Windows/x64: 1 Regression on 8/24/2026 9:06:03 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | x64
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [64a29867c5c5c8e0babc9e31313cf3b6e50f0193](https://github.com/dotnet/sdk/commit/64a29867c5c5c8e0babc9e31313cf3b6e50f0193)
Compare | [c545c4e3f9e2ebc10c6b3e2626187e18569adcf2](https://github.com/dotnet/sdk/commit/c545c4e3f9e2ebc10c6b3e2626187e18569adcf2)
Diff | [Diff](https://github.com/dotnet/sdk/compare/64a29867c5c5c8e0babc9e31313cf3b6e50f0193...c545c4e3f9e2ebc10c6b3e2626187e18569adcf2)
Configs | CompilationMode:tiered, RunKind:imagesharp

### Regressions in SixLabors.ImageSharp.Benchmarks.Codecs.Jpeg.DecodeJpeg

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

• [JpegBaseline400 - 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 'SixLabors.ImageSharp.Benchmarks.Codecs.Jpeg.DecodeJpeg*'
```

### SixLabors.ImageSharp.Benchmarks.Codecs.Jpeg.DecodeJpeg.JpegBaseline400

#### 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/ImageSharp/Codecs/Jpeg/DecodeJpeg.cs at the linked benchmark source, then run the provided benchmarks_ci.py command using the stated filter. Compare the baseline and test commits and use the profiling and benchmarking workflow documents to investigate the 1.48x result. Done means the regression is explained and the benchmark result is validated after the relevant change.

Written by the indexing model from the issue text.