[Perf] Linux/arm64: 27 Improvements on 4/18/2026 5:58:25 PM +00:00

Open

#71,830 opened on Apr 23, 2026 0 comments 0 reactions 0 assignees View on GitHub

ampere arch-arm64 branch-refs/heads/main compilationmode-tiered kind-micro os-linux perf-improvement 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 | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [38af44d518908fa536ef289d62c73706769d18ab](https://github.com/dotnet/runtime/commit/38af44d518908fa536ef289d62c73706769d18ab)
Compare | [b70c35ed8a2e7ae0d91de76f4f5d26c2e7d2c6cd](https://github.com/dotnet/runtime/commit/b70c35ed8a2e7ae0d91de76f4f5d26c2e7d2c6cd)
Diff | [Diff](https://github.com/dotnet/runtime/compare/38af44d518908fa536ef289d62c73706769d18ab...b70c35ed8a2e7ae0d91de76f4f5d26c2e7d2c6cd)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Collections.Perf_LengthBucketsFrozenDictionary

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

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

| 338.76 ns | 252.49 ns | 0.75 | 0.56 | False | | |
|

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

| 23.76 μs | 17.75 μs | 0.75 | 0.51 | False | | |
|

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

| 24.15 μs | 18.10 μs | 0.75 | 0.49 | False | | |
|

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

| 329.75 ns | 251.76 ns | 0.76 | 0.53 | False | | |
|

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

| 2.23 μs | 1.80 μs | 0.81 | 0.52 | False | | |
|

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

| 518.85 μs | 439.12 μs | 0.85 | 0.56 | False | | |
|

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

| 2.09 μs | 1.79 μs | 0.86 | 0.48 | 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.Collections.Perf_LengthBucketsFrozenDictionary*'
```

### System.Collections.Perf_LengthBucketsFrozenDictionary.ToFrozenDictionary(Count: 10, ItemsPerBucket: 5)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_LengthBucketsFrozenDictionary.ToFrozenDictionary(Count: 1000, ItemsPerBucket: 1)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_LengthBucketsFrozenDictionary.ToFrozenDictionary(Count: 1000, ItemsPerBucket: 5)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_LengthBucketsFrozenDictionary.ToFrozenDictionary(Count: 10, ItemsPerBucket: 1)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_LengthBucketsFrozenDictionary.ToFrozenDictionary(Count: 100, ItemsPerBucket: 1)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_LengthBucketsFrozenDictionary.ToFrozenDictionary(Count: 10000, ItemsPerBucket: 5)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_LengthBucketsFrozenDictionary.ToFrozenDictionary(Count: 100, ItemsPerBucket: 5)

#### 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.Collections.CreateAddAndClear<String>

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

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

| 5.11 μs | 3.05 μs | 0.60 | 0.52 | False | | |
|

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

| 3.67 μs | 2.31 μs | 0.63 | 0.51 | False | | |
|

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

| 5.12 μs | 3.72 μs | 0.73 | 0.48 | False | | |
|

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

| 4.75 μs | 3.54 μs | 0.75 | 0.52 | False | | |
|

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

| 4.58 μs | 3.70 μs | 0.81 | 0.49 | 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.Collections.CreateAddAndClear*'
```

### System.Collections.CreateAddAndClear<String>.Stack(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CreateAddAndClear<String>.Array(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CreateAddAndClear<String>.Queue(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CreateAddAndClear<String>.List(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CreateAddAndClear<String>.ICollection(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

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

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

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

| 4.06 μs | 2.69 μs | 0.66 | 0.54 | False | | |
|

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

| 3.90 μs | 2.85 μs | 0.73 | 0.55 | False | | |
|

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

| 3.36 μs | 2.68 μs | 0.80 | 0.54 | False | | |
|

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

| 3.72 μs | 3.03 μs | 0.81 | 0.51 | 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.Collections.AddGivenSize*'
```

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.IterateForEach<String>

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]()

| 4.95 μs | 3.43 μs | 0.69 | 0.45 | False | | |
|

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

| 6.09 μs | 4.77 μs | 0.78 | 0.46 | False | | |
|

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

| 6.77 μs | 6.11 μs | 0.90 | 0.36 | 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.Collections.IterateForEach*'
```

### System.Collections.IterateForEach<String>.ConcurrentBag(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.IterateForEach<String>.SortedSet(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.CreateAddAndRemove<String>

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

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

| 5.65 μs | 4.51 μs | 0.80 | 0.49 | False | | |
|

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

| 6.29 μs | 5.88 μs | 0.93 | 0.40 | 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.Collections.CreateAddAndRemove*'
```

### System.Collections.CreateAddAndRemove<String>.Stack(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CreateAddAndRemove<String>.Queue(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.Perf_DefaultFrozenDictionary

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

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

| 1.37 ms | 1.29 ms | 0.94 | 0.02 | 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.Collections.Perf_DefaultFrozenDictionary*'
```

### System.Collections.Perf_DefaultFrozenDictionary.ToFrozenDictionary(Count: 10000)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.CtorFromCollection<String>

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

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

| 55.19 μs | 52.03 μs | 0.94 | 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
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.CtorFromCollection*'
```

### System.Collections.CtorFromCollection<String>.FrozenDictionaryOptimized(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.CtorFromCollectionNonGeneric<String>

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

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

| 4.95 μs | 3.95 μs | 0.80 | 0.52 | 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.Collections.CtorFromCollectionNonGeneric*'
```

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

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.Perf_SubstringFrozenDictionary

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

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

| 1.77 ms | 1.64 ms | 0.93 | 0.01 | 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.Collections.Perf_SubstringFrozenDictionary*'
```

### System.Collections.Perf_SubstringFrozenDictionary.ToFrozenDictionary(Count: 10000)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in PerfLabTests.DelegatePerf

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

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

| 343.75 ms | 179.77 ms | 0.52 | 0.50 | 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 'PerfLabTests.DelegatePerf*'
```

### PerfLabTests.DelegatePerf.MulticastDelegateCombineInvoke

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.Perf_SingleCharFrozenDictionary

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

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

| 6.17 μs | 5.86 μs | 0.95 | 0.27 | 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.Collections.Perf_SingleCharFrozenDictionary*'
```

### System.Collections.Perf_SingleCharFrozenDictionary.ToFrozenDictionary(Count: 100)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

[Perf] Linux/arm64: 27 Improvements on 4/18/2026 5:58:25 PM +00:00

Description

Contributor guide

Assessment

Get new issues in your inbox