[Perf] Linux/arm64: 1 Regression on 8/21/2026 7:14:37 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [6ecee4dd49b5ab3f97c467372fee3affa17a5a9b](https://github.com/dotnet/runtime/commit/6ecee4dd49b5ab3f97c467372fee3affa17a5a9b)
Compare | [129ada6e62fb997068c4151f69ecb3e39fb6f4be](https://github.com/dotnet/runtime/commit/129ada6e62fb997068c4151f69ecb3e39fb6f4be)
Diff | [Diff](https://github.com/dotnet/runtime/compare/6ecee4dd49b5ab3f97c467372fee3affa17a5a9b...129ada6e62fb997068c4151f69ecb3e39fb6f4be)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in System.Collections.CopyTo<String>

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

• [Array - 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 'System.Collections.CopyTo*'
```

### System.Collections.CopyTo<String>.Array(Size: 2048)

#### 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 System.Collections.CopyTo benchmark source in performance/src/benchmarks/micro/libraries/System.Collections/CopyTo.cs#L33-L34 and the linked regression report. Reproduce it with scripts/benchmarks_ci.py using the provided filter, then compare the baseline and test results for the arm64 AzureLinux configuration. Done means the reported regression is understood and addressed, with the benchmark result verified.

Written by the indexing model from the issue text.