[Perf] Linux/arm64: 16 Regressions on 6/28/2026 5:31:55 PM +00:00

Open

#75,500 opened on Jul 2, 2026 1 comment 0 reactions 0 assignees View on GitHub

arch-arm64 branch-refs/heads/main compilationmode-tiered kind-micro Look Again os-linux perf-regression r2rtype-r2r_interpreter runkind-micro runtime-coreclr untriaged

Dominant language: No language data
Stars: 10
Forks: 8
PR merge metrics: PR metrics pending

Description

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [19a803d6d08a846656d1116d0f4e7001e792d5a1](https://github.com/dotnet/runtime/commit/19a803d6d08a846656d1116d0f4e7001e792d5a1)
Compare | [9cb68cd564627fd80fb9077731093f46d1e0e667](https://github.com/dotnet/runtime/commit/9cb68cd564627fd80fb9077731093f46d1e0e667)
Diff | [Diff](https://github.com/dotnet/runtime/compare/19a803d6d08a846656d1116d0f4e7001e792d5a1...9cb68cd564627fd80fb9077731093f46d1e0e667)
Configs | CompilationMode:tiered, R2RType:r2r_interpreter, RunKind:micro

### Regressions in System.Threading.Channels.Tests.SpscUnboundedChannelPerfTests

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

[TryWriteThenTryRead - Duration of single invocation]()
[ADX - Test Multi Config Graph]()

| 56.18 ns | 63.94 ns | 1.14 | 0.12 | False | | |
|

[WriteAsyncThenReadAsync - Duration of single invocation]()
[ADX - Test Multi Config Graph]()

| 88.77 ns | 96.86 ns | 1.09 | 0.13 | False | | |

![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.Threading.Channels.Tests.SpscUnboundedChannelPerfTests*'
```

### System.Threading.Channels.Tests.SpscUnboundedChannelPerfTests.TryWriteThenTryRead

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Threading.Channels.Tests.SpscUnboundedChannelPerfTests.WriteAsyncThenReadAsync

#### 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

### Regressions in System.Threading.Tests.Perf_SemaphoreSlim

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

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

| 75.37 ns | 82.79 ns | 1.10 | 0.13 | False | | |
|

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

| 84.28 ns | 90.15 ns | 1.07 | 0.33 | False | | |

![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.Threading.Tests.Perf_SemaphoreSlim*'
```

### System.Threading.Tests.Perf_SemaphoreSlim.ReleaseWaitAsync

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Threading.Tests.Perf_SemaphoreSlim.ReleaseWait

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.IO.Pipelines.Tests.Perf_Pipe

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

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

| 4.63 ms | 4.96 ms | 1.07 | 0.09 | False | | |
|

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

| 1.34 ms | 1.43 ms | 1.07 | 0.15 | False | | |

![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.IO.Pipelines.Tests.Perf_Pipe*'
```

### System.IO.Pipelines.Tests.Perf_Pipe.ReadAsync

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Pipelines.Tests.Perf_Pipe.SyncReadAsyncWithCancellationToken

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.IndexerSet<Int32>

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

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

| 24.77 μs | 28.98 μs | 1.17 | 0.07 | 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 'System.Collections.IndexerSet*'
```

### System.Collections.IndexerSet<Int32>.ConcurrentDictionary(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

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

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

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

| 112.84 μs | 121.49 μs | 1.08 | 0.06 | 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 'System.Collections.AddGivenSize*'
```

### System.Collections.AddGivenSize<String>.ConcurrentDictionary(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Threading.Channels.Tests.BoundedChannelPerfTests

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

[TryWriteThenTryRead - Duration of single invocation]()
[ADX - Test Multi Config Graph]()

| 102.24 ns | 115.26 ns | 1.13 | 0.10 | 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 'System.Threading.Channels.Tests.BoundedChannelPerfTests*'
```

### System.Threading.Channels.Tests.BoundedChannelPerfTests.TryWriteThenTryRead

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Threading.Tests.Perf_Monitor

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

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

| 32.23 ns | 35.39 ns | 1.10 | 0.55 | 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 'System.Threading.Tests.Perf_Monitor*'
```

### System.Threading.Tests.Perf_Monitor.EnterExit

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Diagnostics.Perf_Process

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

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

| 182.49 ns | 200.50 ns | 1.10 | 0.09 | 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 'System.Diagnostics.Perf_Process*'
```

### System.Diagnostics.Perf_Process.GetCurrentProcess

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.Concurrent.Count<Int32>

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

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

| 80.64 ns | 89.05 ns | 1.10 | 0.12 | 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 'System.Collections.Concurrent.Count*'
```

### System.Collections.Concurrent.Count<Int32>.Bag(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

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

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

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

| 43.56 μs | 49.09 μs | 1.13 | 0.15 | 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 'System.Collections.IndexerSet*'
```

### System.Collections.IndexerSet<String>.ConcurrentDictionary(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

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

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

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

| 156.36 μs | 166.00 μs | 1.06 | 0.06 | 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 'System.Collections.TryAddDefaultSize*'
```

### System.Collections.TryAddDefaultSize<String>.ConcurrentDictionary(Count: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.Concurrent.Count<String>

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

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

| 76.05 ns | 87.69 ns | 1.15 | 0.10 | 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 'System.Collections.Concurrent.Count*'
```

### System.Collections.Concurrent.Count<String>.Bag(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in Microsoft.AspNetCore.Server.Kestrel.Performance.PipeThroughputBenchmark

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

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

| 1.10 μs | 1.16 μs | 1.05 | 0.07 | 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 'Microsoft.AspNetCore.Server.Kestrel.Performance.PipeThroughputBenchmark*'
```

### Microsoft.AspNetCore.Server.Kestrel.Performance.PipeThroughputBenchmark.Parse_SequentialAsync(Length: 4096, Chunks: 16)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the benchmark workflow docs and run benchmarks_ci.py using the listed System.Threading.Channels.Tests.SpscUnboundedChannelPerfTests and System.Threading.Tests.Perf_SemaphoreSlim filters. Read the named benchmark source src/benchmarks/micro/libraries/System.Threading/Perf.SemaphoreSlim.cs and the reported System.IO.Pipelines benchmark source. Done means explaining and resolving the ARM64 AzureLinux regressions, then confirming the affected benchmarks no longer regress.

Written by the indexing model from the issue text.

Assessment

Tech stack: git, linux, python
Domain: performance, testing
Issue type: Bug
Difficulty: 4/5
Estimated time: 3-5 days
Activity status: Quiet
Clarity: Needs clarification
Newbie friendliness: 38/100

[Perf] Linux/arm64: 16 Regressions on 6/28/2026 5:31:55 PM +00:00

Description

Contributor guide

Research direction

Assessment

Get new issues in your inbox