dotnet / dotnet/perf-autofiling-issues

[Perf] Linux/arm64: 44 Improvements on 5/7/2026 11:58:29 AM +00:00

Open
#72,966 2 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 metrics pending

Description

### Run Information

Name | Value
-- | --
Architecture | arm64
OS | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [4da638db14b248048f4efc07c8166c1f521b208e](https://github.com/dotnet/runtime/commit/4da638db14b248048f4efc07c8166c1f521b208e)
Compare | [c3caf4d3163c429efbdefe7756a71702e9e6c982](https://github.com/dotnet/runtime/commit/c3caf4d3163c429efbdefe7756a71702e9e6c982)
Diff | [Diff](https://github.com/dotnet/runtime/compare/4da638db14b248048f4efc07c8166c1f521b208e...c3caf4d3163c429efbdefe7756a71702e9e6c982)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Text.Json.Tests.Perf_Reader

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

  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1086.17 ns | 897.63 ns | 0.83 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1103.28 ns | 918.82 ns | 0.83 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1.85 μs | 1.63 μs | 0.88 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 91.68 ns | 81.16 ns | 0.89 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 106.15 ns | 93.99 ns | 0.89 | 0.00 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1.87 μs | 1.68 μs | 0.90 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 6.21 μs | 5.63 μs | 0.91 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 90.45 ns | 82.49 ns | 0.91 | 0.02 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1.85 μs | 1.69 μs | 0.91 | 0.02 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 76.25 ns | 69.81 ns | 0.92 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.92 μs | 2.68 μs | 0.92 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1.87 μs | 1.72 μs | 0.92 | 0.02 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.94 μs | 2.73 μs | 0.93 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 543.32 ns | 504.83 ns | 0.93 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 559.59 ns | 521.05 ns | 0.93 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.62 μs | 2.44 μs | 0.93 | 0.00 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 11.46 μs | 10.86 μs | 0.95 | 0.01 | False | | |

![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![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.Text.Json.Tests.Perf_Reader*'
```

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: LotsOfStrings)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: True, TestCase: LotsOfStrings)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: False, TestCase: LotsOfStrings)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: False, TestCase: HelloWorld)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: HelloWorld)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: LotsOfStrings)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: BroadTree)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: True, TestCase: HelloWorld)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: LotsOfNumbers)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: HelloWorld)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: DeepTree)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: True, TestCase: LotsOfNumbers)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: True, TestCase: DeepTree)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: True, TestCase: Json400B)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: True, TestCase: Json400B)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: False, TestCase: LotsOfNumbers)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: False, TestCase: BroadTree)

#### 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 | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [4da638db14b248048f4efc07c8166c1f521b208e](https://github.com/dotnet/runtime/commit/4da638db14b248048f4efc07c8166c1f521b208e)
Compare | [c3caf4d3163c429efbdefe7756a71702e9e6c982](https://github.com/dotnet/runtime/commit/c3caf4d3163c429efbdefe7756a71702e9e6c982)
Diff | [Diff](https://github.com/dotnet/runtime/compare/4da638db14b248048f4efc07c8166c1f521b208e...c3caf4d3163c429efbdefe7756a71702e9e6c982)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Tests.Perf_Version

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

  • [TryParse4 - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 62.49 ns | 48.83 ns | 0.78 | 0.03 | False | | |
|
  • [TryParse3 - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 49.07 ns | 38.60 ns | 0.79 | 0.03 | False | | |
|
  • [TryParse2 - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 34.89 ns | 27.83 ns | 0.80 | 0.03 | False | | |
|
  • [Parse3 - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 49.20 ns | 41.37 ns | 0.84 | 0.03 | False | | |
|
  • [Parse2 - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 34.64 ns | 29.21 ns | 0.84 | 0.07 | False | | |
|
  • [Parse4 - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 62.43 ns | 53.35 ns | 0.85 | 0.04 | False | | |

![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.Tests.Perf_Version*'
```

### System.Tests.Perf_Version.TryParse4

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Version.TryParse3

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Version.TryParse2

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Version.Parse3

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Version.Parse2

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Version.Parse4

#### 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 | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [4da638db14b248048f4efc07c8166c1f521b208e](https://github.com/dotnet/runtime/commit/4da638db14b248048f4efc07c8166c1f521b208e)
Compare | [c3caf4d3163c429efbdefe7756a71702e9e6c982](https://github.com/dotnet/runtime/commit/c3caf4d3163c429efbdefe7756a71702e9e6c982)
Diff | [Diff](https://github.com/dotnet/runtime/compare/4da638db14b248048f4efc07c8166c1f521b208e...c3caf4d3163c429efbdefe7756a71702e9e6c982)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>

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

  • [Divide_Scalar - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 5.22 μs | 1.56 μs | 0.30 | 0.00 | False | | |
|
  • [Divide_Scalar - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 216.79 ns | 68.30 ns | 0.32 | 0.00 | False | | |
|
  • [IndexOfMax - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 51.56 ns | 41.32 ns | 0.80 | 0.01 | False | | |
|
  • [IndexOfMax - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1.22 μs | 1.03 μs | 0.85 | 0.01 | 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.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives*'
```

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.Divide_Scalar(BufferLength: 3079)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.Divide_Scalar(BufferLength: 128)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.IndexOfMax(BufferLength: 128)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.IndexOfMax(BufferLength: 3079)

#### 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 | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [4da638db14b248048f4efc07c8166c1f521b208e](https://github.com/dotnet/runtime/commit/4da638db14b248048f4efc07c8166c1f521b208e)
Compare | [c3caf4d3163c429efbdefe7756a71702e9e6c982](https://github.com/dotnet/runtime/commit/c3caf4d3163c429efbdefe7756a71702e9e6c982)
Diff | [Diff](https://github.com/dotnet/runtime/compare/4da638db14b248048f4efc07c8166c1f521b208e...c3caf4d3163c429efbdefe7756a71702e9e6c982)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Text.Json.Tests.Perf_Depth

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

  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.32 μs | 2.04 μs | 0.88 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.27 μs | 1.99 μs | 0.88 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 20.52 μs | 18.18 μs | 0.89 | 0.02 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 76.42 ns | 71.92 ns | 0.94 | 0.01 | 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.Text.Json.Tests.Perf_Depth*'
```

### System.Text.Json.Tests.Perf_Depth.ReadSpanEmptyLoop(Depth: 65)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Depth.ReadSpanEmptyLoop(Depth: 64)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Depth.ReadSpanEmptyLoop(Depth: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Depth.ReadSpanEmptyLoop(Depth: 1)

#### 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 | ubuntu 22.04
Queue | AmpereUbuntu
Baseline | [4da638db14b248048f4efc07c8166c1f521b208e](https://github.com/dotnet/runtime/commit/4da638db14b248048f4efc07c8166c1f521b208e)
Compare | [c3caf4d3163c429efbdefe7756a71702e9e6c982](https://github.com/dotnet/runtime/commit/c3caf4d3163c429efbdefe7756a71702e9e6c982)
Diff | [Diff](https://github.com/dotnet/runtime/compare/4da638db14b248048f4efc07c8166c1f521b208e...c3caf4d3163c429efbdefe7756a71702e9e6c982)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Tests.Perf_Type

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

  • [GetType_Resolvers - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 275.35 ns | 233.22 ns | 0.85 | 0.38 | False | | |
|
  • [GetType_Resolvers - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 296.89 ns | 267.82 ns | 0.90 | 0.36 | False | | |
|
  • [GetType_Resolvers - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 131.09 ns | 124.20 ns | 0.95 | 0.48 | 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.Tests.Perf_Type*'
```

### System.Tests.Perf_Type.GetType_Resolvers(input: typeof(System.Int32[]))

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Type.GetType_Resolvers(input: typeof(System.Tests.Nested))

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_Type.GetType_Resolvers(input: typeof(int))

#### 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)

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.