[Perf] Windows/x86: 104 Improvements on 3/7/2026 5:41:53 AM +00:00

Open

#69,809 opened on Mar 10, 2026 3 comments 0 reactions 0 assignees View on GitHub

arch-x86 branch-refs/heads/main kind-micro os-windows perf-improvement runtime-coreclr untriaged

Dominant language: No language data
Stars: 10
Forks: 8
PR merge metrics: PR metrics pending

Description

### Run Information

Name | Value
-- | --
Architecture | x86
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [703be0833b619fab3b970175f0d391c097d1e17b](https://github.com/dotnet/runtime/commit/703be0833b619fab3b970175f0d391c097d1e17b)
Compare | [28baa459661f164b55d6409f9816178ced46146f](https://github.com/dotnet/runtime/commit/28baa459661f164b55d6409f9816178ced46146f)
Diff | [Diff](https://github.com/dotnet/runtime/compare/703be0833b619fab3b970175f0d391c097d1e17b...28baa459661f164b55d6409f9816178ced46146f)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in PerfLabTests.DelegatePerf

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

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

| 14.98 ms | 324.55 μs | 0.02 | 0.03 | False | | |
|

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

| 712.53 ms | 21.72 ms | 0.03 | 0.32 | False | | |
|

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

| 71.41 ms | 1.92 ms | 0.03 | 0.12 | False | | |

[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.DelegatePerf*'
```

### PerfLabTests.DelegatePerf.DelegateInvoke

#### ETL Files

#### Histogram

#### JIT Disasms

### PerfLabTests.DelegatePerf.MulticastDelegateInvoke(length: 1000)

#### ETL Files

#### Histogram

#### JIT Disasms

### PerfLabTests.DelegatePerf.MulticastDelegateInvoke(length: 100)

#### 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 | x86
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [64efb69573d1aa99fa89a96de18ba3247b2bb75e](https://github.com/dotnet/runtime/commit/64efb69573d1aa99fa89a96de18ba3247b2bb75e)
Compare | [28baa459661f164b55d6409f9816178ced46146f](https://github.com/dotnet/runtime/commit/28baa459661f164b55d6409f9816178ced46146f)
Diff | [Diff](https://github.com/dotnet/runtime/compare/64efb69573d1aa99fa89a96de18ba3247b2bb75e...28baa459661f164b55d6409f9816178ced46146f)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in ArrayDeAbstraction

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

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

| 2.15 μs | 1.80 μs | 0.84 | 0.31 | False | | |

[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 'ArrayDeAbstraction*'
```

### ArrayDeAbstraction.foreach_opaque_array_via_interface_in_loop

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in MicroBenchmarks.Serializers.Xml_ToStream<Location>

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

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

| 1.68 μs | 1.06 μs | 0.63 | 0.06 | False | | |
|

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

| 1.08 μs | 583.43 ns | 0.54 | 0.11 | False | | |

[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 'MicroBenchmarks.Serializers.Xml_ToStream*'
```

### MicroBenchmarks.Serializers.Xml_ToStream<Location>.DataContractSerializer_

#### ETL Files

#### Histogram

#### JIT Disasms

### MicroBenchmarks.Serializers.Xml_ToStream<Location>.DataContractSerializer_BinaryXml_

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Linq.Tests.Perf_Enumerable

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

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

| 639.30 ns | 326.75 ns | 0.51 | 0.59 | False | | |
|

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

| 1.21 μs | 495.72 ns | 0.41 | 0.43 | False | | |
|

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

| 1.05 μs | 522.13 ns | 0.50 | 0.40 | False | | |
|

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

| 1.02 μs | 731.45 ns | 0.72 | 0.33 | False | | |
|

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

| 671.89 ns | 431.08 ns | 0.64 | 0.33 | False | | |
|

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

| 1.10 μs | 517.52 ns | 0.47 | 0.56 | False | | |
|

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

| 754.30 ns | 341.47 ns | 0.45 | 0.50 | False | | |
|

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

| 982.98 ns | 496.87 ns | 0.51 | 0.41 | False | | |

[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.Linq.Tests.Perf_Enumerable*'
```

### System.Linq.Tests.Perf_Enumerable.Select(input: Array)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Linq.Tests.Perf_Enumerable.Take_All(input: IEnumerable)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Linq.Tests.Perf_Enumerable.Select(input: IList)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Linq.Tests.Perf_Enumerable.LastWithPredicate_FirstElementMatches(input: IOrderedEnumerable)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Linq.Tests.Perf_Enumerable.Range

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Linq.Tests.Perf_Enumerable.Skip_One(input: IEnumerable)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Linq.Tests.Perf_Enumerable.SkipHalfTakeHalf(input: IEnumerable)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Linq.Tests.Perf_Enumerable.Select(input: IEnumerable)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in MicroBenchmarks.Serializers.Xml_FromStream<XmlElement>

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

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

| 623.15 ns | 412.51 ns | 0.66 | 0.09 | False | | |

[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 'MicroBenchmarks.Serializers.Xml_FromStream*'
```

### MicroBenchmarks.Serializers.Xml_FromStream<XmlElement>.DataContractSerializer_BinaryXml_

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in HardwareIntrinsics.RayTracer.SoA

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

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

| 15.09 ms | 6.58 ms | 0.44 | 0.30 | False | | |

[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 'HardwareIntrinsics.RayTracer.SoA*'
```

### HardwareIntrinsics.RayTracer.SoA.Render

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

Name | Value
-- | --
Architecture | x86
OS | Windows 10.0.22631
Queue | ViperWindows
Baseline | [703be0833b619fab3b970175f0d391c097d1e17b](https://github.com/dotnet/runtime/commit/703be0833b619fab3b970175f0d391c097d1e17b)
Compare | [8151eb1c17763d564c5283f9a0db96ebb9856390](https://github.com/dotnet/runtime/commit/8151eb1c17763d564c5283f9a0db96ebb9856390)
Diff | [Diff](https://github.com/dotnet/runtime/compare/703be0833b619fab3b970175f0d391c097d1e17b...8151eb1c17763d564c5283f9a0db96ebb9856390)
Configs | CompilationMode:tiered, RunKind:micro

### Improvements in System.Collections.Perf_SingleCharFrozenDictionary

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

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

| 128.03 μs | 71.35 μs | 0.56 | 0.03 | False | | |
|

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

| 158.92 ns | 77.15 ns | 0.49 | 0.02 | False | | |
|

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

| 12.86 μs | 7.15 μs | 0.56 | 0.02 | False | | |
|

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

| 1.29 μs | 719.93 ns | 0.56 | 0.02 | False | | |
|

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

| 1.58 μs | 757.78 ns | 0.48 | 0.06 | False | | |
|

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

| 15.96 μs | 7.64 μs | 0.48 | 0.02 | False | | |
|

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

| 157.58 μs | 75.92 μs | 0.48 | 0.02 | False | | |
|

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

| 130.89 ns | 73.17 ns | 0.56 | 0.02 | False | | |

[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_SingleCharFrozenDictionary*'
```

### System.Collections.Perf_SingleCharFrozenDictionary.TryGetValue_False_FrozenDictionary(Count: 10000)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_SingleCharFrozenDictionary.TryGetValue_True_FrozenDictionary(Count: 10)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_SingleCharFrozenDictionary.TryGetValue_False_FrozenDictionary(Count: 1000)

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

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

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_SingleCharFrozenDictionary.TryGetValue_True_FrozenDictionary(Count: 1000)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_SingleCharFrozenDictionary.TryGetValue_True_FrozenDictionary(Count: 10000)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_SingleCharFrozenDictionary.TryGetValue_False_FrozenDictionary(Count: 10)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in MicroBenchmarks.Serializers.Xml_ToStream<MyEventsListerViewModel>

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

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

| 400.39 μs | 234.65 μs | 0.59 | 0.08 | False | | |

[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 'MicroBenchmarks.Serializers.Xml_ToStream*'
```

### MicroBenchmarks.Serializers.Xml_ToStream<MyEventsListerViewModel>.DataContractSerializer_BinaryXml_

#### 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
-- | -- | -- | -- | -- | -- | -- | -- | --
|

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

| 134.87 ns | 74.78 ns | 0.55 | 0.05 | False | | |
|

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

| 159.14 ns | 77.24 ns | 0.49 | 0.03 | False | | |
|

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

| 2.36 μs | 1.52 μs | 0.64 | 0.02 | False | | |
|

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

| 23.50 μs | 15.19 μs | 0.65 | 0.01 | False | | |
|

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

| 1.74 μs | 1.15 μs | 0.66 | 0.01 | False | | |

[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_DefaultFrozenDictionary*'
```

### System.Collections.Perf_DefaultFrozenDictionary.TryGetValue_False_FrozenDictionary(Count: 10)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_DefaultFrozenDictionary.TryGetValue_True_FrozenDictionary(Count: 10)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_DefaultFrozenDictionary.TryGetValue_True_FrozenDictionary(Count: 100)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_DefaultFrozenDictionary.TryGetValue_True_FrozenDictionary(Count: 1000)

#### ETL Files

#### Histogram

#### JIT Disasms

### System.Collections.Perf_DefaultFrozenDictionary.TryGetValue_False_FrozenDictionary(Count: 100)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in MicroBenchmarks.Serializers.Json_ToStream<Location>

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

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

| 2.00 μs | 1.19 μs | 0.60 | 0.17 | False | | |

[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 'MicroBenchmarks.Serializers.Json_ToStream*'
```

### MicroBenchmarks.Serializers.Json_ToStream<Location>.DataContractJsonSerializer_

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

---

### Run Information

### Improvements in System.Collections.CreateAddAndRemove<Int32>

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

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

| 87.17 μs | 50.12 μs | 0.57 | 0.11 | False | | |

[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.CreateAddAndRemove*'
```

### System.Collections.CreateAddAndRemove<Int32>.SortedDictionary(Size: 512)

#### ETL Files

#### Histogram

#### JIT Disasms

### Docs

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

[Perf] Windows/x86: 104 Improvements on 3/7/2026 5:41:53 AM +00:00

Description

Contributor guide

Assessment

Get new issues in your inbox