[Perf] Linux/arm64: 7 Improvements on 2/12/2025 6:39:37 AM +00:00

[performanceautofiler]

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	cc74292fb04d03b4fc91df200fab3da50654d64a
Compare	33612803be925e3ae2f8d8e66a068591a265d904
Diff	Diff
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_NonFullyQualifiedNames - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.68 μs	1.53 μs	0.91	0.07	False
GetType_FullyQualifiedNames - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.81 μs	1.61 μs	0.89	0.06	False
GetType_InvalidName - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.07 μs	964.64 ns	0.90	0.01	False
GetType_NonFullyQualifiedNames - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.83 μs	1.68 μs	0.92	0.05	False
GetType_FullyQualifiedNames - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	2.06 μs	1.93 μs	0.94	0.06	False

Test Report

Repro

General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md

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_NonFullyQualifiedNames(input: "System.Int32")

ETL Files

Histogram

JIT Disasms

System.Tests.Perf_Type.GetType_FullyQualifiedNames(input: typeof(int))

ETL Files

Histogram

JIT Disasms

System.Tests.Perf_Type.GetType_InvalidName

ETL Files

Histogram

JIT Disasms

System.Tests.Perf_Type.GetType_NonFullyQualifiedNames(input: "System.Int32[]")

ETL Files

Histogram

JIT Disasms

System.Tests.Perf_Type.GetType_FullyQualifiedNames(input: typeof(System.Int32[,]))

ETL Files

Histogram

JIT Disasms

Docs

Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	cc74292fb04d03b4fc91df200fab3da50654d64a
Compare	33612803be925e3ae2f8d8e66a068591a265d904
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Improvements in System.Memory.ReadOnlySequence

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Slice_Start - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	6.67 ns	0.78 ns	0.12	0.56	False
Slice_Start - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	5.57 ns	0.83 ns	0.15	0.65	False

Test Report

Repro

General Docs link: https://github.com/dotnet/performance/blob/main/docs/benchmarking-workflow-dotnet-runtime.md

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Memory.ReadOnlySequence*'

System.Memory.ReadOnlySequence.Slice_Start(Segment: Multiple)

ETL Files

Histogram

JIT Disasms

System.Memory.ReadOnlySequence.Slice_Start(Segment: Single)

ETL Files

Histogram

JIT Disasms

Docs

Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository

[AndyAyersMS]

Probably dotnet/runtime#112060