[Perf] Linux/arm64: 12 Improvements on 11/1/2024 1:56:26 AM

[performanceautofiler]

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	d2a2a79b2eaf66d68a02f35917907bd9fbe97ea7
Compare	357b09da5abffaeb67d64c928e15293ca2a2de5e
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Improvements in System.Numerics.Tests.Perf_VectorOf<Int64>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
MultiplyBenchmark - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.81 ns	0.05 ns	0.02	0.40	True
DotBenchmark - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.87 ns	0.00 ns	0.00	0.41	False
MultiplyOperatorBenchmark - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.86 ns	0.01 ns	0.00	0.42	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.Numerics.Tests.Perf_VectorOf<Int64>*'

System.Numerics.Tests.Perf_VectorOf<Int64>.MultiplyBenchmark

ETL Files

Histogram

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<Int64>.DotBenchmark

ETL Files

Histogram

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<Int64>.MultiplyOperatorBenchmark

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	d2a2a79b2eaf66d68a02f35917907bd9fbe97ea7
Compare	357b09da5abffaeb67d64c928e15293ca2a2de5e
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Improvements in System.Numerics.Tests.Perf_VectorOf<UInt64>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
MultiplyBenchmark - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.92 ns	0.04 ns	0.02	0.41	False
DotBenchmark - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	2.00 ns	0.00 ns	0.00	0.46	False
MultiplyOperatorBenchmark - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.91 ns	0.02 ns	0.01	0.39	True

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.Numerics.Tests.Perf_VectorOf<UInt64>*'

System.Numerics.Tests.Perf_VectorOf<UInt64>.MultiplyBenchmark

ETL Files

Histogram

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt64>.DotBenchmark

ETL Files

Histogram

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt64>.MultiplyOperatorBenchmark

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	d2a2a79b2eaf66d68a02f35917907bd9fbe97ea7
Compare	357b09da5abffaeb67d64c928e15293ca2a2de5e
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Improvements in System.Linq.Tests.Perf_Enumerable

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
WhereAny_LastElementMatches - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	132.62 ns	100.43 ns	0.76	0.46	False
OrderBy - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	2.28 μs	1.98 μs	0.87	0.01	False
FirstWithPredicate_LastElementMatches - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.94 μs	1.64 μs	0.85	0.02	False
WhereFirst_LastElementMatches - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	124.92 ns	92.03 ns	0.74	0.39	False
AnyWithPredicate_LastElementMatches - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.93 μs	1.68 μs	0.87	0.02	False
OrderByDescending - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	2.96 μs	2.57 μs	0.87	0.01	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.Linq.Tests.Perf_Enumerable*'

System.Linq.Tests.Perf_Enumerable.WhereAny_LastElementMatches(input: Array)

ETL Files

Histogram

JIT Disasms

System.Linq.Tests.Perf_Enumerable.OrderBy(input: IEnumerable)

ETL Files

Histogram

JIT Disasms

System.Linq.Tests.Perf_Enumerable.FirstWithPredicate_LastElementMatches(input: IOrderedEnumerable)

ETL Files

Histogram

JIT Disasms

System.Linq.Tests.Perf_Enumerable.WhereFirst_LastElementMatches(input: Array)

ETL Files

Histogram

JIT Disasms

System.Linq.Tests.Perf_Enumerable.AnyWithPredicate_LastElementMatches(input: IOrderedEnumerable)

ETL Files

Histogram

JIT Disasms

System.Linq.Tests.Perf_Enumerable.OrderByDescending(input: IEnumerable)

ETL Files

Histogram

JIT Disasms

Docs

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

[LoopedBard3]

Vector Improvements caused by: dotnet/runtime#106594, other improvements caused by: dotnet/runtime#109407