[Perf] Linux/arm64: 26 Regressions on 7/29/2026 4:04:54 AM +00:00

オープン

#77,458 2026年8月6日に作成コメント 4 件リアクション 0 件担当者 0 名 GitHub で見る

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

主要言語: 言語のデータがありません
スター: 10
フォーク: 8
PR マージ指標: PR 指標を取得中

説明

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [975877333c98d9d08bad66670d74accc6047be34](https://github.com/dotnet/runtime/commit/975877333c98d9d08bad66670d74accc6047be34)
Compare | [3dad2c9f5b81d9263c01399d8b28c32b979c2da5](https://github.com/dotnet/runtime/commit/3dad2c9f5b81d9263c01399d8b28c32b979c2da5)
Diff | [Diff](https://github.com/dotnet/runtime/compare/975877333c98d9d08bad66670d74accc6047be34...3dad2c9f5b81d9263c01399d8b28c32b979c2da5)
Configs | CompilationMode:tiered, RunKind:micro

### Regressions in Microsoft.Extensions.DependencyInjection.GetService

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

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

| 5.33 ns | 7.86 ns | 1.47 | 0.22 | False | | |
|

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

| 44.84 ns | 54.68 ns | 1.22 | 0.23 | False | | |
|

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

| 68.19 ns | 79.21 ns | 1.16 | 0.20 | False | | |
|

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

| 21.06 ns | 24.37 ns | 1.16 | 0.14 | 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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'Microsoft.Extensions.DependencyInjection.GetService*'
```

### Microsoft.Extensions.DependencyInjection.GetService.NoDI

#### ETL Files

#### Histogram

#### JIT Disasms

### Microsoft.Extensions.DependencyInjection.GetService.ServiceScope

#### ETL Files

#### Histogram

#### JIT Disasms

### Microsoft.Extensions.DependencyInjection.GetService.Scoped

#### ETL Files

#### Histogram

#### JIT Disasms

### Microsoft.Extensions.DependencyInjection.GetService.Transient

#### 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.Collections.CtorFromCollectionNonGeneric<Int32>

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

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

| 6.30 μs | 7.93 μs | 1.26 | 0.19 | False | | |
|

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

| 8.38 μs | 9.75 μs | 1.16 | 0.10 | False | | |
|

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

| 24.73 μs | 27.26 μs | 1.10 | 0.18 | 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.Collections.CtorFromCollectionNonGeneric*'
```

### System.Collections.CtorFromCollectionNonGeneric<Int32>.ArrayList(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CtorFromCollectionNonGeneric<Int32>.Stack(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CtorFromCollectionNonGeneric<Int32>.Hashtable(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.TryGetValueFalse<String, String>

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

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

| 343.29 μs | 424.70 μs | 1.24 | 0.24 | False | | |
|

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

| 323.32 μs | 376.22 μs | 1.16 | 0.22 | False | | |
|

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

| 258.68 μs | 297.47 μs | 1.15 | 0.24 | 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.Collections.TryGetValueFalse*'
```

### System.Collections.TryGetValueFalse<String, String>.SortedList(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.TryGetValueFalse<String, String>.ImmutableSortedDictionary(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.TryGetValueFalse<String, String>.SortedDictionary(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.Tests.Add_Remove_SteadyState<Int32>

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

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

| 18.02 ns | 46.96 ns | 2.61 | 0.53 | False | | |
|

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

| 31.26 ns | 39.91 ns | 1.28 | 0.22 | False | | |
|

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

| 11.76 ns | 14.76 ns | 1.26 | 0.16 | 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.Collections.Tests.Add_Remove_SteadyState*'
```

### System.Collections.Tests.Add_Remove_SteadyState<Int32>.ConcurrentBag(Count: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Tests.Add_Remove_SteadyState<Int32>.ImmutableQueue(Count: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Tests.Add_Remove_SteadyState<Int32>.ImmutableStack(Count: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in PerfLabTests.CastingPerf2.CastingPerf

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

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

| 720.04 μs | 957.73 μs | 1.33 | 0.18 | False | | |
|

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

| 729.38 μs | 961.27 μs | 1.32 | 0.24 | False | | |
|

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

| 963.56 μs | 1190.55 μs | 1.24 | 0.20 | 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.CastingPerf2.CastingPerf*'
```

### PerfLabTests.CastingPerf2.CastingPerf.ObjScalarValueType

#### ETL Files

#### Histogram

#### JIT Disasms

### PerfLabTests.CastingPerf2.CastingPerf.ObjInt

#### ETL Files

#### Histogram

#### JIT Disasms

### PerfLabTests.CastingPerf2.CastingPerf.ObjObjrefValueType

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Text.Json.Node.Tests.Perf_Create

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

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

| 660.22 ns | 757.28 ns | 1.15 | 0.19 | False | | |
|

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

| 1.10 μs | 1.24 μs | 1.13 | 0.25 | 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.Text.Json.Node.Tests.Perf_Create*'
```

### System.Text.Json.Node.Tests.Perf_Create.Create_JsonString

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Node.Tests.Perf_Create.Create_JsonArray

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Text.Json.Tests.Perf_Ctor

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

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

| 21.88 ns | 26.67 ns | 1.22 | 0.47 | False | | |
|

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

| 23.96 ns | 28.58 ns | 1.19 | 0.31 | 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.Text.Json.Tests.Perf_Ctor*'
```

### System.Text.Json.Tests.Perf_Ctor.Ctor(Formatted: True, SkipValidation: False)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Ctor.Ctor(Formatted: False, SkipValidation: False)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.Perf_Frozen<NotKnownComparable>

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

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

| 43.54 ns | 49.75 ns | 1.14 | 0.19 | False | | |
|

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

| 91.10 ns | 102.90 ns | 1.13 | 0.17 | 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.Collections.Perf_Frozen*'
```

### System.Collections.Perf_Frozen<NotKnownComparable>.ToFrozenDictionary(Count: 4)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_Frozen<NotKnownComparable>.ToFrozenSet(Count: 4)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Security.Cryptography.Tests.Perf_CryptoConfig

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

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

| 491.42 ns | 568.21 ns | 1.16 | 0.17 | 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.Security.Cryptography.Tests.Perf_CryptoConfig*'
```

### System.Security.Cryptography.Tests.Perf_CryptoConfig.CreateFromName(name: "RSA")

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Net.NetworkInformation.Tests.PhysicalAddressTests

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

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

| 15.56 ns | 17.79 ns | 1.14 | 0.13 | 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.Net.NetworkInformation.Tests.PhysicalAddressTests*'
```

### System.Net.NetworkInformation.Tests.PhysicalAddressTests.PAMedium

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.Perf_Frozen<ReferenceType>

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

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

| 97.35 ns | 104.67 ns | 1.08 | 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 'System.Collections.Perf_Frozen*'
```

### System.Collections.Perf_Frozen<ReferenceType>.ToFrozenDictionary(Count: 4)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Globalization.Tests.Perf_DateTimeCultureInfo

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

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

| 252.66 ns | 279.15 ns | 1.10 | 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.Globalization.Tests.Perf_DateTimeCultureInfo*'
```

### System.Globalization.Tests.Perf_DateTimeCultureInfo.ToStringHebrewIsrael

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

コントリビューションガイド

このリポジトリのコントリビューションガイドは索引されていません

調査の方向性

Start with the linked baseline/compare diff and the performance report, then review GetServiceBenchmark.cs and CtorFromCollectionNonGeneric.cs. Run the documented benchmarks_ci.py command and compare the ARM64 AzureLinux results against the supplied baseline. Done means identifying and addressing the reported regressions, with the benchmark results confirming the change.

索引モデルが issue の本文から書いたものです。

評価

技術スタック: azure, csharp, linux
領域: operating-systems, performance, testing-qa
issue の種類: バグ
難易度: 4/5
見積もり時間: 3〜5日
活発さ: 静か
明瞭さ: 説明が足りない
初心者へのやさしさ: 30/100

[Perf] Linux/arm64: 26 Regressions on 7/29/2026 4:04:54 AM +00:00

説明

コントリビューションガイド

調査の方向性

評価

新しい issue をメールで受け取る