[Perf] Linux/arm64: 58 Regressions on 8/4/2026 8:53:30 PM +00:00

Open

#77,464 opened on Aug 6, 2026 3 comments 0 reactions 0 assignees View on GitHub

arch-arm64 bdn-update branch-refs/heads/main compilationmode-tiered kind-micro Look Again os-linux perf-regression 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 | azurelinux 3.0
Queue | CobaltAzureLinux
Baseline | [2dd5330804f9d4f49599d82c90a67892d4dd8af0](https://github.com/dotnet/runtime/commit/2dd5330804f9d4f49599d82c90a67892d4dd8af0)
Compare | [173e6b40e02cd1652102c09e07bd4cf7320b3d4d](https://github.com/dotnet/runtime/commit/173e6b40e02cd1652102c09e07bd4cf7320b3d4d)
Diff | [Diff](https://github.com/dotnet/runtime/compare/2dd5330804f9d4f49599d82c90a67892d4dd8af0...173e6b40e02cd1652102c09e07bd4cf7320b3d4d)
Configs | CompilationMode:tiered, RunKind:micro

### Regressions in SeekUnroll

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

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

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

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

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

[Test - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config 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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'SeekUnroll*'
```

### SeekUnroll.Test(boxedIndex: 1)

#### ETL Files

#### Histogram

#### JIT Disasms

### SeekUnroll.Test(boxedIndex: 19)

#### ETL Files

#### Histogram

#### JIT Disasms

### SeekUnroll.Test(boxedIndex: 3)

#### ETL Files

#### Histogram

#### JIT Disasms

### SeekUnroll.Test(boxedIndex: 27)

#### ETL Files

#### Histogram

#### JIT Disasms

### SeekUnroll.Test(boxedIndex: 11)

#### 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 Burgers

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

[Test1 - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 331.53 ms | 3264.53 ms | 9.85 | 0.15 | False | | |
|

[Test0 - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 381.26 ms | 3548.15 ms | 9.31 | 0.13 | False | | |
|

[Test3 - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 430.68 ms | 1367.75 ms | 3.18 | 0.12 | False | | |
|

[Test2 - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 330.47 ms | 895.40 ms | 2.71 | 0.24 | 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 'Burgers*'
```

### Burgers.Test1

#### ETL Files

#### Histogram

#### JIT Disasms

### Burgers.Test0

#### ETL Files

#### Histogram

#### JIT Disasms

### Burgers.Test3

#### ETL Files

#### Histogram

#### JIT Disasms

### Burgers.Test2

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in LinqBenchmarks

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

[Count00ForX - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 53.98 ms | 722.59 ms | 13.39 | 0.23 | False | | |
|

[Where00ForX - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 176.21 ms | 866.39 ms | 4.92 | 0.19 | False | | |
|

[Where01ForX - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config Graph]()

| 122.33 ms | 441.31 ms | 3.61 | 0.22 | False | | |
|

[CountBy00LookupX - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config 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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'LinqBenchmarks*'
```

### LinqBenchmarks.Count00ForX

#### ETL Files

#### Histogram

#### JIT Disasms

### LinqBenchmarks.Where00ForX

#### ETL Files

#### Histogram

#### JIT Disasms

### LinqBenchmarks.Where01ForX

#### ETL Files

#### Histogram

#### JIT Disasms

### LinqBenchmarks.CountBy00LookupX

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.IO.Tests.Perf_StreamWriter

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

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

| 126.51 ms | 178.60 ms | 1.41 | 0.28 | False | | |
|

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

| 128.53 ms | 179.34 ms | 1.40 | 0.24 | False | | |
|

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

| 140.93 ms | 181.60 ms | 1.29 | 0.25 | 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.IO.Tests.Perf_StreamWriter*'
```

### System.IO.Tests.Perf_StreamWriter.WriteCharArray(writeLength: 2)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_StreamWriter.WriteString(writeLength: 2)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.IO.Tests.Perf_StreamWriter.WritePartialCharArray(writeLength: 2)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Tests.Perf_GC<Byte>

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

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

| 2.46 μs | 3.08 μs | 1.25 | 0.29 | False | | |
|

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

| 90.19 ns | 106.17 ns | 1.18 | 0.49 | False | | |
|

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

| 2.48 μs | 2.89 μs | 1.17 | 0.31 | 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.Tests.Perf_GC*'
```

### System.Tests.Perf_GC<Byte>.AllocateUninitializedArray(length: 10000, pinned: True)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_GC<Byte>.AllocateArray(length: 1000, pinned: False)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Tests.Perf_GC<Byte>.AllocateArray(length: 10000, pinned: True)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in SciMark2.kernel

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

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

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

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

| 725.31 ms | 986.44 ms | 1.36 | 0.19 | 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 'SciMark2.kernel*'
```

### SciMark2.kernel.benchSparseMult

#### ETL Files

#### Histogram

#### JIT Disasms

### SciMark2.kernel.benchSOR

#### ETL Files

#### Histogram

#### JIT Disasms

### SciMark2.kernel.benchMonteCarlo

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in System.Collections.CopyTo<Int32>

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

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

| 162.33 ns | 195.01 ns | 1.20 | 0.21 | False | | |
|

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

| 172.51 ns | 196.23 ns | 1.14 | 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.CopyTo*'
```

### System.Collections.CopyTo<Int32>.Span(Size: 2048)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.CopyTo<Int32>.Memory(Size: 2048)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in Benchstone.MDBenchI.MDLogicArray

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

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

| 515.23 ms | 621.43 ms | 1.21 | 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 'Benchstone.MDBenchI.MDLogicArray*'
```

### Benchstone.MDBenchI.MDLogicArray.Test

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in Benchstone.BenchF.BenchMrk

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

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

| 190.77 ms | 339.71 ms | 1.78 | 0.12 | 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 'Benchstone.BenchF.BenchMrk*'
```

### Benchstone.BenchF.BenchMrk.Test

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in Benchstone.MDBenchI.MDArray2

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

[Test - Duration of single invocation]()
📝 - [Benchmark Source]()
[ADX - Test Multi Config 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 'Benchstone.MDBenchI.MDArray2*'
```

### Benchstone.MDBenchI.MDArray2.Test

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in Benchstone.BenchF.Lorenz

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

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

| 392.60 ms | 992.32 ms | 2.53 | 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 'Benchstone.BenchF.Lorenz*'
```

### Benchstone.BenchF.Lorenz.Test

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Regressions in Benchstone.BenchI.LogicArray

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

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

| 393.68 ms | 549.37 ms | 1.40 | 0.19 | 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 'Benchstone.BenchI.LogicArray*'
```

### Benchstone.BenchI.LogicArray.Test

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

Contributor guide

No contributing guide indexed for this repository

Research direction

Start with the linked benchmark sources SeekUnroll.cs lines 59-80 and Burgers.cs lines 185-189, then review the baseline and compare commits in the dotnet/runtime diff. Reproduce the ARM64 Azure Linux regression with the documented benchmarks_ci.py command and inspect the linked reports. Done means identifying and addressing the reported performance regressions, with benchmark results showing whether they are resolved.

Written by the indexing model from the issue text.

Assessment

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

[Perf] Linux/arm64: 58 Regressions on 8/4/2026 8:53:30 PM +00:00

Description

Contributor guide

Research direction

Assessment

Get new issues in your inbox