dotnet / dotnet/perf-autofiling-issues

[Perf] Linux/arm64: 55 Regressions on 6/24/2026 6:25:29 AM +00:00

Open
#75,490 2 comments 0 reactions 0 assignees View on GitHub
ampere arch-arm64 branch-refs/heads/main compilationmode-tiered kind-micro os-linux perf-regression runkind-micro runtime-coreclr
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 | [f4430a4ba762ef82841ff7a8c340ed8854b2bca9](https://github.com/dotnet/runtime/commit/f4430a4ba762ef82841ff7a8c340ed8854b2bca9)
Compare | [37d45de0d2f8d77b150163a96052354486c20a25](https://github.com/dotnet/runtime/commit/37d45de0d2f8d77b150163a96052354486c20a25)
Diff | [Diff](https://github.com/dotnet/runtime/compare/f4430a4ba762ef82841ff7a8c340ed8854b2bca9...37d45de0d2f8d77b150163a96052354486c20a25)
Configs | CompilationMode:tiered, RunKind:micro

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

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

  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.19 μs | 3.24 μs | 1.48 | 0.12 | False | | |
|
  • [ReadReturnBytes - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.55 μs | 3.37 μs | 1.32 | 0.18 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 92.92 ns | 119.26 ns | 1.28 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 85.62 ns | 109.60 ns | 1.28 | 0.04 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 1.69 μs | 2.14 μs | 1.27 | 0.12 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.47 μs | 3.12 μs | 1.26 | 0.13 | False | | |
|
  • [ReadReturnBytes - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 860.56 ns | 1019.73 ns | 1.18 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 782.52 ns | 921.92 ns | 1.18 | 0.01 | False | | |
|
  • [ReadReturnBytes - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 95.09 ns | 112.00 ns | 1.18 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 803.41 ns | 941.17 ns | 1.17 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 67.39 μs | 78.90 μs | 1.17 | 0.01 | False | | |
|
  • [ReadReturnBytes - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 7.58 μs | 8.83 μs | 1.16 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 6.84 μs | 7.92 μs | 1.16 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 6.87 μs | 7.91 μs | 1.15 | 0.01 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 67.87 μs | 77.78 μs | 1.15 | 0.01 | False | | |
|
  • [ReadMultiSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 3.19 μs | 3.64 μs | 1.14 | 0.13 | False | | |
|
  • [ReadReturnBytes - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 76.88 μs | 87.57 μs | 1.14 | 0.01 | False | | |
|
  • [ReadMultiSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 993.63 ns | 1130.14 ns | 1.14 | 0.01 | False | | |
|
  • [ReadMultiSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 75.59 μs | 85.84 μs | 1.14 | 0.02 | False | | |
|
  • [ReadSpanEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 10.86 μs | 12.27 μs | 1.13 | 0.01 | False | | |
|
  • [ReadMultiSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 8.12 μs | 9.09 μs | 1.12 | 0.01 | False | | |
|
  • [ReadReturnBytes - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.10 μs | 2.34 μs | 1.12 | 0.03 | False | | |
|
  • [ReadMultiSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 152.24 ns | 169.92 ns | 1.12 | 0.01 | False | | |
|
  • [ReadSingleSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 11.09 μs | 12.30 μs | 1.11 | 0.02 | False | | |
|
  • [ReadMultiSpanSequenceEmptyLoop - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 2.03 μs | 2.23 μs | 1.10 | 0.01 | False | | |
|
  • [ReadReturnBytes - Duration of single invocation]()
  • 📝 - [Benchmark Source]()
  • [ADX - Test Multi Config Graph]()
| 11.95 μs | 13.11 μs | 1.10 | 0.00 | False | | |

![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![graph]()
![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.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: LotsOfNumbers)

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### 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.ReadSpanEmptyLoop(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: False, TestCase: LotsOfNumbers)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadReturnBytes(IsDataCompact: False, TestCase: Json400B)

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: Json40KB)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadReturnBytes(IsDataCompact: False, TestCase: Json4KB)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: False, TestCase: Json4KB)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSingleSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: Json4KB)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: False, TestCase: Json40KB)

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadReturnBytes(IsDataCompact: False, TestCase: Json40KB)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadMultiSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: Json400B)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadMultiSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: Json40KB)

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadMultiSpanSequenceEmptyLoop(IsDataCompact: False, TestCase: Json4KB)

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Text.Json.Tests.Perf_Reader.ReadReturnBytes(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)

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the Run Information and compare the listed baseline and compare commits, then inspect the linked System.Text.Json benchmark reports. Read the benchmark source in performance/src/benchmarks/micro/libraries/System.Text.Json/Utf8JsonReader/Perf.Reader.cs at the referenced lines and check whether the reported ARM64 regressions reproduce. Done means the cause is identified and the affected benchmark results are addressed.

Written by the indexing model from the issue text.

Assessment

Tech stack
linux
Domain
performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.