[Perf] Linux/arm64: 28 Improvements on 2/15/2026 2:14:43 AM -08:00

Open

#69,046 opened on Feb 19, 2026 0 comments 0 reactions 0 assignees View on GitHub

ampere arch-arm64 branch-refs/heads/main kind-micro os-linux perf-improvement 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 | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [1be81201324528909e18b8c08511d24e277e70cc](https://github.com/dotnet/runtime/commit/1be81201324528909e18b8c08511d24e277e70cc)
Compare | [253a5c5911d9447ac19864a215408440228da2ed](https://github.com/dotnet/runtime/commit/253a5c5911d9447ac19864a215408440228da2ed)
Diff | [Diff](https://github.com/dotnet/runtime/compare/1be81201324528909e18b8c08511d24e277e70cc...253a5c5911d9447ac19864a215408440228da2ed)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Threading.Tests.Perf_SemaphoreSlim

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

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

| 2.91 μs | 2.56 μs | 0.88 | 0.38 | False | | |
|

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

| 4.02 μs | 2.90 μs | 0.72 | 0.49 | False | | |
|

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

| 2.99 μs | 2.51 μs | 0.84 | 0.51 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Threading.Tests.Perf_SemaphoreSlim*'
```

### System.Threading.Tests.Perf_SemaphoreSlim.ReleaseWaitAsync_WithTimeout

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Threading.Tests.Perf_SemaphoreSlim.ReleaseWaitAsync_WithCancellationTokenAndTimeout

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Threading.Tests.Perf_SemaphoreSlim.ReleaseWaitAsync_WithCancellationToken

#### 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

### Improvements in System.IO.Tests.Perf_FileStream

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]()

| 1.20 ms | 647.33 μs | 0.54 | 0.72 | False | | |
|

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

| 54.22 ms | 26.49 ms | 0.49 | 0.61 | False | | |
|

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

| 110.64 ms | 63.62 ms | 0.58 | 0.67 | False | | |
|

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

| 1.79 ms | 1.44 ms | 0.80 | 0.68 | False | | |
|

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

| 1.17 ms | 789.39 μs | 0.67 | 0.70 | False | | |
|

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

| 93.83 ms | 72.91 ms | 0.78 | 0.73 | False | | |
|

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

| 157.37 μs | 137.17 μs | 0.87 | 0.55 | False | | |

![graph]()
![graph]()
![graph]()
![graph]()
![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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.IO.Tests.Perf_FileStream*'
```

### System.IO.Tests.Perf_FileStream.ReadAsync(fileSize: 1048576, userBufferSize: 4096, options: Asynchronous)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_FileStream.ReadAsync_NoBuffering(fileSize: 104857600, userBufferSize: 16384, options: None)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_FileStream.ReadAsync(fileSize: 104857600, userBufferSize: 4096, options: Asynchronous)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_FileStream.ReadAsync(fileSize: 1048576, userBufferSize: 512, options: Asynchronous)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_FileStream.ReadAsync(fileSize: 1048576, userBufferSize: 4096, options: None)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_FileStream.ReadAsync(fileSize: 104857600, userBufferSize: 4096, options: None)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_FileStream.CopyToFileAsync(fileSize: 1024, options: None)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Buffers.Tests.RentReturnArrayPoolTests<Object>

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

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

| 1.34 μs | 1.10 μs | 0.82 | 0.42 | False | | |
|

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

| 1.81 μs | 1.13 μs | 0.63 | 0.68 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Buffers.Tests.RentReturnArrayPoolTests*'
```

### System.Buffers.Tests.RentReturnArrayPoolTests<Object>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: True, UseSharedPool: False)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Buffers.Tests.RentReturnArrayPoolTests<Object>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: True, UseSharedPool: True)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Buffers.Tests.RentReturnArrayPoolTests<Byte>

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

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

| 4.25 μs | 3.21 μs | 0.76 | 0.10 | False | | |
|

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

| 4.33 μs | 2.50 μs | 0.58 | 0.11 | False | | |
|

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

| 1.27 μs | 1.15 μs | 0.91 | 0.46 | False | | |
|

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

| 1.66 μs | 1.16 μs | 0.70 | 0.67 | False | | |

![graph]()
![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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Buffers.Tests.RentReturnArrayPoolTests*'
```

### System.Buffers.Tests.RentReturnArrayPoolTests<Byte>.SingleSerial(RentalSize: 4096, ManipulateArray: True, Async: True, UseSharedPool: False)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Buffers.Tests.RentReturnArrayPoolTests<Byte>.SingleSerial(RentalSize: 4096, ManipulateArray: True, Async: True, UseSharedPool: True)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Buffers.Tests.RentReturnArrayPoolTests<Byte>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: True, UseSharedPool: False)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Buffers.Tests.RentReturnArrayPoolTests<Byte>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: True, UseSharedPool: True)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Net.Security.Tests.SslStreamTests

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

[DefaultMutualHandshakeContextIPv4Async - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 4.97 ms | 3.98 ms | 0.80 | 0.10 | False | | |
|

[DefaultHandshakeIPv4Async - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 4.40 ms | 4.06 ms | 0.92 | 0.12 | False | | |
|

[DefaultHandshakeIPv6Async - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 4.46 ms | 4.07 ms | 0.91 | 0.12 | False | | |
|

[DefaultMutualHandshakeContextIPv6Async - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 5.02 ms | 3.93 ms | 0.78 | 0.11 | False | | |
|

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

| 4.40 ms | 3.91 ms | 0.89 | 0.11 | False | | |

![graph]()
![graph]()
![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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Net.Security.Tests.SslStreamTests*'
```

### System.Net.Security.Tests.SslStreamTests.DefaultMutualHandshakeContextIPv4Async

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Net.Security.Tests.SslStreamTests.DefaultHandshakeIPv4Async

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Net.Security.Tests.SslStreamTests.DefaultHandshakeIPv6Async

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Net.Security.Tests.SslStreamTests.DefaultMutualHandshakeContextIPv6Async

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Net.Security.Tests.SslStreamTests.DefaultHandshakePipeAsync

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.IO.Tests.Perf_RandomAccess

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]()

| 48.20 ms | 25.48 ms | 0.53 | 0.55 | False | | |
|

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

| 1.04 ms | 775.36 μs | 0.75 | 0.67 | False | | |
|

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

| 1.12 ms | 727.53 μs | 0.65 | 0.66 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.IO.Tests.Perf_RandomAccess*'
```

### System.IO.Tests.Perf_RandomAccess.ReadAsync(fileSize: 104857600, bufferSize: 16384, options: Asynchronous)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_RandomAccess.ReadAsync(fileSize: 1048576, bufferSize: 4096, options: None)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_RandomAccess.ReadAsync(fileSize: 1048576, bufferSize: 4096, options: Asynchronous)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in BenchmarksGame.Fasta_1

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

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

| 701.26 μs | 609.62 μs | 0.87 | 0.37 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'BenchmarksGame.Fasta_1*'
```

### BenchmarksGame.Fasta_1.RunBench

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.IO.Tests.Perf_File

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

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

| 140.34 μs | 113.07 μs | 0.81 | 0.39 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.IO.Tests.Perf_File*'
```

### System.IO.Tests.Perf_File.WriteAllTextAsync(size: 10000)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

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

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

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

| 3.24 μs | 2.03 μs | 0.63 | 0.75 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'Microsoft.AspNetCore.Server.Kestrel.Performance.PipeThroughputBenchmark*'
```

### Microsoft.AspNetCore.Server.Kestrel.Performance.PipeThroughputBenchmark.Parse_ParallelAsync(Length: 128, Chunks: 16)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Threading.Tasks.ValueTaskPerfTest

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

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

| 2.02 μs | 1.69 μs | 0.84 | 0.41 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Threading.Tasks.ValueTaskPerfTest*'
```

### System.Threading.Tasks.ValueTaskPerfTest.CreateAndAwait_FromYieldingAsyncMethod

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the linked benchmark sources, including src/benchmarks/micro/libraries/System.Threading/Perf.SemaphoreSlim.cs and src/benchmarks/micro/libraries/System.IO.FileSystem/Perf.FileStream.cs, then run the benchmarks_ci.py command from the issue. Compare the arm64 Ubuntu results with the linked reports; the issue does not define a code change or completion condition, so triage is needed before implementation.

Written by the indexing model from the issue text.

Assessment

Tech stack: csharp, python
Domain: performance, testing-qa
Issue type: Bug
Difficulty: 5/5
Estimated time: Over a week
Activity status: Stale
Clarity: Needs clarification
Newbie friendliness: 15/100

[Perf] Linux/arm64: 28 Improvements on 2/15/2026 2:14:43 AM -08:00

Description

Contributor guide

Research direction

Assessment

Get new issues in your inbox