[Perf] Linux/arm64: 5 Regressions on 8/24/2026 11:05:53 PM +00:00

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [51874e577c504e758a38f19bfb9999aa1ad0594a](https://github.com/dotnet/runtime/commit/51874e577c504e758a38f19bfb9999aa1ad0594a)
Compare | [df846d69e1666f9c6060f2487975b568f3ed93bb](https://github.com/dotnet/runtime/commit/df846d69e1666f9c6060f2487975b568f3ed93bb)
Diff | [Diff](https://github.com/dotnet/runtime/compare/51874e577c504e758a38f19bfb9999aa1ad0594a...df846d69e1666f9c6060f2487975b568f3ed93bb)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in PerfLabTests.CastingPerf

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

• [ObjFooIsObj2 - Duration of single invocation]()
• 📝 - [Benchmark Source]()
• [ADX - Test Multi Config Graph]()

| 338.94 μs | 395.96 μs | 1.17 | 0.13 | False | | |
|

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

| 338.62 μs | 388.50 μs | 1.15 | 0.14 | False | | |
|

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

| 350.80 μs | 392.84 μs | 1.12 | 0.13 | 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 'PerfLabTests.CastingPerf*'
```

### PerfLabTests.CastingPerf.ObjFooIsObj2

#### ETL Files

#### Histogram

#### JIT Disasms

### PerfLabTests.CastingPerf.IFooFooIsIFoo

#### ETL Files

#### Histogram

#### JIT Disasms

### PerfLabTests.CastingPerf.ObjFooIsObj

#### 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)

---

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [51874e577c504e758a38f19bfb9999aa1ad0594a](https://github.com/dotnet/runtime/commit/51874e577c504e758a38f19bfb9999aa1ad0594a)
Compare | [df846d69e1666f9c6060f2487975b568f3ed93bb](https://github.com/dotnet/runtime/commit/df846d69e1666f9c6060f2487975b568f3ed93bb)
Diff | [Diff](https://github.com/dotnet/runtime/compare/51874e577c504e758a38f19bfb9999aa1ad0594a...df846d69e1666f9c6060f2487975b568f3ed93bb)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in PerfLabTests.LowLevelPerf

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

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

| 190.87 μs | 238.61 μs | 1.25 | 0.16 | False | | |

![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 'PerfLabTests.LowLevelPerf*'
```

### PerfLabTests.LowLevelPerf.GenericClassWithSTringGenericInstanceMethod

#### 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)

---

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [51874e577c504e758a38f19bfb9999aa1ad0594a](https://github.com/dotnet/runtime/commit/51874e577c504e758a38f19bfb9999aa1ad0594a)
Compare | [df846d69e1666f9c6060f2487975b568f3ed93bb](https://github.com/dotnet/runtime/commit/df846d69e1666f9c6060f2487975b568f3ed93bb)
Diff | [Diff](https://github.com/dotnet/runtime/compare/51874e577c504e758a38f19bfb9999aa1ad0594a...df846d69e1666f9c6060f2487975b568f3ed93bb)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in PerfLabTests.CastingPerf2.CastingPerf

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

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

| 381.72 μs | 477.19 μs | 1.25 | 0.20 | False | | |

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

### PerfLabTests.CastingPerf2.CastingPerf.FooObjIsNull

#### 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 sources src/benchmarks/micro/runtime/perflab/CastingPerf.cs, LowLevelPerf.cs, and CastingPerf2.cs at the linked lines, then review the baseline/compare commits and regression reports. Run benchmarks_ci.py with the listed filters and compare the reported ARM64 results; done means the regressions are reproduced and their cause or disposition is established.

Written by the indexing model from the issue text.