[Perf] Linux/arm64: 3 Regressions on 7/10/2026 4:24:12 AM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [6c5849144dc8a23aa0760080d6f7784fed60da37](https://github.com/dotnet/runtime/commit/6c5849144dc8a23aa0760080d6f7784fed60da37)
Compare | [1d85f408cee362c3b693e8b2ccba114b976b317d](https://github.com/dotnet/runtime/commit/1d85f408cee362c3b693e8b2ccba114b976b317d)
Diff | [Diff](https://github.com/dotnet/runtime/compare/6c5849144dc8a23aa0760080d6f7784fed60da37...1d85f408cee362c3b693e8b2ccba114b976b317d)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in System.Tests.Perf_Int32

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

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

| 21.98 ns | 24.83 ns | 1.13 | 0.11 | False | | |
|

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

| 29.85 ns | 32.20 ns | 1.08 | 0.10 | False | | |
|

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

| 30.44 ns | 32.15 ns | 1.06 | 0.10 | False | | |

![graph]()
![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.Tests.Perf_Int32*'
```

### System.Tests.Perf_Int32.Parse(value: "4")

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Int32.ParseHex(value: "80000000")

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Int32.ParseHex(value: "7FFFFFFF")

#### 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 linked System.Tests.Perf_Int32 benchmark source, performance report, and the baseline/compare commits; then run the provided benchmarks_ci.py command for the Perf_Int32 filter. Compare the three Parse and ParseHex regressions on Linux/arm64 under the listed configuration, and consider the work complete when the regressions are reproduced, explained, and their outcome is recorded.

Written by the indexing model from the issue text.