[Perf] Linux/arm64: 1 Regression on 7/13/2026 3:15:10 PM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [8df98c4d2d17bcba98193afa9ae2e72908921f86](https://github.com/dotnet/runtime/commit/8df98c4d2d17bcba98193afa9ae2e72908921f86)
Compare | [db330a0e1ac8c7950a07271b2a5d8ac1830243bf](https://github.com/dotnet/runtime/commit/db330a0e1ac8c7950a07271b2a5d8ac1830243bf)
Diff | [Diff](https://github.com/dotnet/runtime/compare/8df98c4d2d17bcba98193afa9ae2e72908921f86...db330a0e1ac8c7950a07271b2a5d8ac1830243bf)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in MicroBenchmarks.Serializers.Json_FromStream<MyEventsListerViewModel>

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

• [SystemTextJson_Reflection_ - Duration of single invocation]()
• [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 'MicroBenchmarks.Serializers.Json_FromStream*'
```

### MicroBenchmarks.Serializers.Json_FromStream<MyEventsListerViewModel>.SystemTextJson_Reflection_

#### 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 scripts/benchmarks_ci.py and the MicroBenchmarks.Serializers.Json_FromStream filter, then compare the linked baseline and test commits and review the benchmark and profiling workflow docs. Done means the arm64 AzureLinux regression is explained and the benchmark no longer shows an unexplained performance degradation.

Written by the indexing model from the issue text.