Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Perf] Linux/arm64: 31 Improvements on 7/6/2023 2:07:14 AM #19758

Closed
performanceautofiler bot opened this issue Jul 11, 2023 · 3 comments
Closed

[Perf] Linux/arm64: 31 Improvements on 7/6/2023 2:07:14 AM #19758

performanceautofiler bot opened this issue Jul 11, 2023 · 3 comments
Labels
ampere arch-arm64 branch-refs/heads/main kind-micro_mono mono-interpreter os-linux perf-improvement PGO Applied if there were any profile guided optimization updates in the observed interval. runtime-mono

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Jul 11, 2023 •
edited
Loading

Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline cd26e63274102c8130109c9ce1301851a550433c
Compare 5d54b08d5fc40d0b1c156f430a487a94c1e34f79
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_Vector4

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
AddOperatorBenchmark - Duration of single invocation 32.69 ns 12.25 ns 0.37 0.36 False
UnitYBenchmark - Duration of single invocation 15.97 ns 2.57 ns 0.16 0.36 False
NegateBenchmark - Duration of single invocation 33.01 ns 8.97 ns 0.27 0.43 False
SubtractFunctionBenchmark - Duration of single invocation 32.94 ns 10.59 ns 0.32 0.37 False
OneBenchmark - Duration of single invocation 14.35 ns 2.74 ns 0.19 0.26 False
SubtractOperatorBenchmark - Duration of single invocation 32.85 ns 12.22 ns 0.37 0.41 False
DistanceSquaredJitOptimizeCanaryBenchmark - Duration of single invocation 54.77 ns 29.34 ns 0.54 0.18 False
SquareRootBenchmark - Duration of single invocation 40.49 ns 32.98 ns 0.81 0.24 False
LerpBenchmark - Duration of single invocation 125.44 ns 79.59 ns 0.63 0.23 False
NegateOperatorBenchmark - Duration of single invocation 33.49 ns 9.55 ns 0.29 0.44 False
UnitWBenchmark - Duration of single invocation 15.77 ns 2.79 ns 0.18 0.37 False
DistanceSquaredBenchmark - Duration of single invocation 57.69 ns 28.59 ns 0.50 0.21 False
DistanceBenchmark - Duration of single invocation 55.22 ns 33.26 ns 0.60 0.19 False
UnitXBenchmark - Duration of single invocation 15.47 ns 3.35 ns 0.22 0.39 False
MultiplyByScalarBenchmark - Duration of single invocation 37.84 ns 29.64 ns 0.78 0.24 False
AddFunctionBenchmark - Duration of single invocation 32.27 ns 12.71 ns 0.39 0.47 False
UnitZBenchmark - Duration of single invocation 15.57 ns 2.58 ns 0.17 0.41 False
CreateFromScalarXYZWBenchmark - Duration of single invocation 15.53 ns 2.89 ns 0.19 0.39 False

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

Repro Steps

Prerequisites (Build files either built locally or downloaded from payload above)

  • Libraries build extracted to runtime/artifacts or build instructions: Libraries README args: -subset libs+libs.tests -rc release -configuration Release -arch $RunArch -framework net8.0
  • CoreCLR product build extracted to runtime/artifacts/bin/coreclr/$RunOS.$RunArch.Release, build instructions: CoreCLR README args: -subset clr+libs -rc release -configuration Release -arch $RunArch -framework net8.0
  • Mono Runtime build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release
  • Dotnet SDK installed for dotnet commands
  • Running commands from the runtime folder

Linux

# Set $RunDir to the runtime directory
RunDir=`pwd`

# Set the OS, arch, and OSId
RunOS='linux'
RunOSId='linux'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir/artifacts/bin/mono/$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
cp $RunDir/artifacts/bin/runtime/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/shared/Microsoft.NETCore.App/8.0.0 -rf
cp $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/dotnet-mono -r
cp $RunDir/artifacts/bin/coreclr/$RunOS.$RunArch.Release/corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun

# Create Core Root
$RunDir/src/tests/build.sh release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir/performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir/performance/scripts/benchmarks_ci.py --csproj $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Vector4*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun" 

# Individual Commands:
# Restore 
dotnet restore $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --packages $RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Vector4*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages

Windows

# Set $RunDir to the runtime directory
$RunDir="FullPathHere"

# Set the OS, arch, and OSId
RunOS='windows'
RunOSId='win'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir\artifacts\bin\mono\$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
xcopy $RunDir\artifacts\bin\runtime\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\shared\Microsoft.NETCore.App\8.0.0\ /e /y
xcopy $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\dotnet-mono\ /e /y
xcopy $RunDir\artifacts\bin\coreclr\$RunOS.$RunArch.Release\corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun /y

# Create Core Root
$RunDir\src\tests\build.cmd release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir\performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir\performance\scripts\benchmarks_ci.py --csproj $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Vector4*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe" 

# Individual Commands:
# Restore 
dotnet restore $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --packages $RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Vector4*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Numerics.Tests.Perf_Vector4.AddOperatorBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/15/2023 1:50:10 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 103.2228849346827 (T) = (0 -12.458113815875036) / Math.Sqrt((9.008345731506253 / (229)) + (0.08923378153985052 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.6853115480487061 = (39.58872255599404 - 12.458113815875036) / 39.58872255599404 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.UnitYBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 2/2/2023 2:17:49 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 52.72449732036053 (T) = (0 -2.8147150660139246) / Math.Sqrt((21.43413487159 / (228)) + (0.0810922439311508 / (3))) is greater than 1.9703772833260038 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (228) + (3) - 2, .975) and 0.8669668633682658 = (21.157999708040357 - 2.8147150660139246) / 21.157999708040357 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.NegateBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/16/2023 12:55:40 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 21.346113858017187 (T) = (0 -11.267865999952422) / Math.Sqrt((10.19035428561103 / (229)) + (4.931441703859244 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.711109710154278 = (39.00396238990889 - 11.267865999952422) / 39.00396238990889 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.SubtractFunctionBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/16/2023 12:55:40 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 99.00878664213096 (T) = (0 -10.945097908459202) / Math.Sqrt((7.9723346771039685 / (229)) + (0.1494687800225699 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.7246454124828868 = (39.749103173300036 - 10.945097908459202) / 39.749103173300036 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.OneBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 10/7/2022 12:48:42 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 37.35052107939728 (T) = (0 -2.9252180908256284) / Math.Sqrt((11.658539625488084 / (229)) + (0.12060471979081769 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.7939900052752037 = (14.199398891948697 - 2.9252180908256284) / 14.199398891948697 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.SubtractOperatorBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/16/2023 12:55:40 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 85.47418086939784 (T) = (0 -12.694212807648896) / Math.Sqrt((8.733497433428234 / (225)) + (0.18356237499193034 / (3))) is greater than 1.9705162426829501 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (225) + (3) - 2, .975) and 0.6804391334033572 = (39.723927847748094 - 12.694212807648896) / 39.723927847748094 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.DistanceSquaredJitOptimizeCanaryBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/6/2023 5:17:50 PM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 50.72931096378595 (T) = (0 -30.07285566685438) / Math.Sqrt((10.914120748177364 / (229)) + (1.279661991423002 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.5373882113184731 = (65.00667817515836 - 30.07285566685438) / 65.00667817515836 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.SquareRootBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 5/14/2023 1:47:19 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 17.916557722703335 (T) = (0 -32.47466243604762) / Math.Sqrt((5.79090330201539 / (229)) + (1.1744019833617347 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.262626222181534 = (44.040978148320534 - 32.47466243604762) / 44.040978148320534 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.LerpBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 10/7/2022 12:48:42 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 23.746795579318928 (T) = (0 -76.40944478754459) / Math.Sqrt((73.04781430657692 / (229)) + (14.81411615889869 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.41608251343032104 = (130.85657913145343 - 76.40944478754459) / 130.85657913145343 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.NegateOperatorBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/19/2023 2:05:48 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 38.93091838718091 (T) = (0 -10.320616127494025) / Math.Sqrt((8.906337886742415 / (222)) + (1.4798911575597675 / (3))) is greater than 1.9706589608356722 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (222) + (3) - 2, .975) and 0.7336885569418485 = (38.753934149350194 - 10.320616127494025) / 38.753934149350194 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.UnitWBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 2/2/2023 2:17:49 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 55.63162863038656 (T) = (0 -2.8534790754927726) / Math.Sqrt((22.557202855735827 / (229)) + (0.0275025732215689 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.8648152496516094 = (21.107995303752418 - 2.8534790754927726) / 21.107995303752418 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.DistanceSquaredBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/6/2023 5:17:50 PM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 19.25462600045903 (T) = (0 -30.35684654257571) / Math.Sqrt((11.636583639319642 / (229)) + (9.553792211543175 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.5329042340007082 = (64.99062666010494 - 30.35684654257571) / 64.99062666010494 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.DistanceBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/6/2023 5:17:50 PM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 97.98867368387421 (T) = (0 -33.08298549503446) / Math.Sqrt((13.15329330236437 / (220)) + (0.15684998241236286 / (3))) is greater than 1.9707562704893595 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (220) + (3) - 2, .975) and 0.4978803999289828 = (65.88666423369128 - 33.08298549503446) / 65.88666423369128 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.UnitXBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 2/2/2023 2:17:49 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 17.439687816732498 (T) = (0 -4.352594956141097) / Math.Sqrt((22.13583228899527 / (228)) + (2.5314923031751464 / (3))) is greater than 1.9703772833260038 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (228) + (3) - 2, .975) and 0.7953574006972157 = (21.269251714796198 - 4.352594956141097) / 21.269251714796198 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.MultiplyByScalarBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 10/7/2022 12:48:42 AM, 5/25/2023 2:16:53 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 24.31994619220104 (T) = (0 -28.849984457010937) / Math.Sqrt((17.338409894019943 / (228)) + (1.3138075610780224 / (3))) is greater than 1.9703772833260038 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (228) + (3) - 2, .975) and 0.37669582699348314 = (46.28556282216532 - 28.849984457010937) / 46.28556282216532 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.AddFunctionBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 6/15/2023 1:50:10 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 45.812164224371614 (T) = (0 -12.1454778585897) / Math.Sqrt((8.139248831554168 / (229)) + (0.9798210900694221 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.694181952807028 = (39.71471916085407 - 12.1454778585897) / 39.71471916085407 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.UnitZBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 2/2/2023 2:17:49 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 20.753285071298812 (T) = (0 -3.2397256626624755) / Math.Sqrt((22.336767309763832 / (229)) + (1.9718277901595835 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.8476872165116743 = (21.27021506970747 - 3.2397256626624755) / 21.27021506970747 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.CreateFromScalarXYZWBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 5/2/2023 12:34:35 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 79.44240130249328 (T) = (0 -3.0751586035376586) / Math.Sqrt((2.7434849404432597 / (229)) + (0.06986697528879425 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.8291059571827495 = (17.99453364694609 - 3.0751586035376586) / 17.99453364694609 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository
@performanceautofiler
Copy link
Author

performanceautofiler bot commented Jul 11, 2023 •
edited
Loading

Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline cd26e63274102c8130109c9ce1301851a550433c
Compare 5d54b08d5fc40d0b1c156f430a487a94c1e34f79
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_Matrix4x4

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
AddBenchmark - Duration of single invocation 382.37 ns 131.48 ns 0.34 0.19 False
CreateFromScalars - Duration of single invocation 102.73 ns 62.91 ns 0.61 0.26 False
SubtractBenchmark - Duration of single invocation 381.38 ns 129.99 ns 0.34 0.23 False
SubtractOperatorBenchmark - Duration of single invocation 381.93 ns 134.24 ns 0.35 0.18 False
AddOperatorBenchmark - Duration of single invocation 382.39 ns 132.78 ns 0.35 0.22 False
CreateFromYawPitchRollBenchmarkBenchmark - Duration of single invocation 277.67 ns 237.80 ns 0.86 0.14 False
CreateScaleFromScalarBenchmark - Duration of single invocation 87.28 ns 24.57 ns 0.28 0.34 False
NegateBenchmark - Duration of single invocation 270.64 ns 106.05 ns 0.39 0.15 False

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

Repro Steps

Prerequisites (Build files either built locally or downloaded from payload above)

  • Libraries build extracted to runtime/artifacts or build instructions: Libraries README args: -subset libs+libs.tests -rc release -configuration Release -arch $RunArch -framework net8.0
  • CoreCLR product build extracted to runtime/artifacts/bin/coreclr/$RunOS.$RunArch.Release, build instructions: CoreCLR README args: -subset clr+libs -rc release -configuration Release -arch $RunArch -framework net8.0
  • Mono Runtime build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release
  • Dotnet SDK installed for dotnet commands
  • Running commands from the runtime folder

Linux

# Set $RunDir to the runtime directory
RunDir=`pwd`

# Set the OS, arch, and OSId
RunOS='linux'
RunOSId='linux'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir/artifacts/bin/mono/$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
cp $RunDir/artifacts/bin/runtime/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/shared/Microsoft.NETCore.App/8.0.0 -rf
cp $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/dotnet-mono -r
cp $RunDir/artifacts/bin/coreclr/$RunOS.$RunArch.Release/corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun

# Create Core Root
$RunDir/src/tests/build.sh release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir/performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir/performance/scripts/benchmarks_ci.py --csproj $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Matrix4x4*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun" 

# Individual Commands:
# Restore 
dotnet restore $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --packages $RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Matrix4x4*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages

Windows

# Set $RunDir to the runtime directory
$RunDir="FullPathHere"

# Set the OS, arch, and OSId
RunOS='windows'
RunOSId='win'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir\artifacts\bin\mono\$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
xcopy $RunDir\artifacts\bin\runtime\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\shared\Microsoft.NETCore.App\8.0.0\ /e /y
xcopy $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\dotnet-mono\ /e /y
xcopy $RunDir\artifacts\bin\coreclr\$RunOS.$RunArch.Release\corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun /y

# Create Core Root
$RunDir\src\tests\build.cmd release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir\performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir\performance\scripts\benchmarks_ci.py --csproj $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Matrix4x4*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe" 

# Individual Commands:
# Restore 
dotnet restore $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --packages $RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Matrix4x4*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Numerics.Tests.Perf_Matrix4x4.AddBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 21.983774984082544 (T) = (0 -129.17867357997704) / Math.Sqrt((16152.621055855068 / (228)) + (64.91923811708166 / (3))) is greater than 1.9703772833260038 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (228) + (3) - 2, .975) and 0.6207255867018077 = (340.5942216260564 - 129.17867357997704) / 340.5942216260564 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Matrix4x4.CreateFromScalars

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 25.8974227851464 (T) = (0 -62.93425236177163) / Math.Sqrt((185.6414228577089 / (228)) + (0.1856444266827104 / (3))) is greater than 1.9703772833260038 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (228) + (3) - 2, .975) and 0.2780639108367855 = (87.1742711113359 - 62.93425236177163) / 87.1742711113359 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Matrix4x4.SubtractBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 22.998240027512793 (T) = (0 -128.9175865686058) / Math.Sqrt((16288.514533662925 / (229)) + (43.3605496951137 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.6226905914087757 = (341.67604526468267 - 128.9175865686058) / 341.67604526468267 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Matrix4x4.SubtractOperatorBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 23.525912878338648 (T) = (0 -132.32334785069278) / Math.Sqrt((17391.838143445897 / (229)) + (7.363470434060159 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.6115363345860422 = (340.6324957308048 - 132.32334785069278) / 340.6324957308048 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Matrix4x4.AddOperatorBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 21.461417405593114 (T) = (0 -129.0554967337298) / Math.Sqrt((17360.616139409718 / (228)) + (61.19434417648479 / (3))) is greater than 1.9703772833260038 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (228) + (3) - 2, .975) and 0.6203417497615806 = (339.925437291788 - 129.0554967337298) / 339.925437291788 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Matrix4x4.CreateFromYawPitchRollBenchmarkBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 3/16/2023 1:27:09 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 14.296773950780649 (T) = (0 -244.28255464971494) / Math.Sqrt((390.08642498045214 / (225)) + (32.60838597662441 / (3))) is greater than 1.9705162426829501 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (225) + (3) - 2, .975) and 0.17202871076274498 = (295.0374702904895 - 244.28255464971494) / 295.0374702904895 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Matrix4x4.CreateScaleFromScalarBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 24.70830282961984 (T) = (0 -25.211721613015403) / Math.Sqrt((1173.766082031399 / (229)) + (0.31875965327935524 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.6915156394214681 = (81.72771405893417 - 25.211721613015403) / 81.72771405893417 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Matrix4x4.NegateBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 1/6/2023 2:21:07 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 20.881568707141117 (T) = (0 -99.35448900199155) / Math.Sqrt((8270.276111859674 / (222)) + (35.90028438900964 / (3))) is greater than 1.9706589608356722 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (222) + (3) - 2, .975) and 0.595879502757632 = (245.85362455001754 - 99.35448900199155) / 245.85362455001754 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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

Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline cd26e63274102c8130109c9ce1301851a550433c
Compare 5d54b08d5fc40d0b1c156f430a487a94c1e34f79
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

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

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
OneBenchmark - Duration of single invocation 45.30 ns 0.94 ns 0.02 0.41 False

graph
Test Report

Repro

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

Repro Steps

Prerequisites (Build files either built locally or downloaded from payload above)

  • Libraries build extracted to runtime/artifacts or build instructions: Libraries README args: -subset libs+libs.tests -rc release -configuration Release -arch $RunArch -framework net8.0
  • CoreCLR product build extracted to runtime/artifacts/bin/coreclr/$RunOS.$RunArch.Release, build instructions: CoreCLR README args: -subset clr+libs -rc release -configuration Release -arch $RunArch -framework net8.0
  • Mono Runtime build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release
  • Dotnet SDK installed for dotnet commands
  • Running commands from the runtime folder

Linux

# Set $RunDir to the runtime directory
RunDir=`pwd`

# Set the OS, arch, and OSId
RunOS='linux'
RunOSId='linux'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir/artifacts/bin/mono/$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
cp $RunDir/artifacts/bin/runtime/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/shared/Microsoft.NETCore.App/8.0.0 -rf
cp $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/dotnet-mono -r
cp $RunDir/artifacts/bin/coreclr/$RunOS.$RunArch.Release/corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun

# Create Core Root
$RunDir/src/tests/build.sh release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir/performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir/performance/scripts/benchmarks_ci.py --csproj $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_VectorOf&lt;Single&gt;*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun" 

# Individual Commands:
# Restore 
dotnet restore $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --packages $RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_VectorOf&lt;Single&gt;*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages

Windows

# Set $RunDir to the runtime directory
$RunDir="FullPathHere"

# Set the OS, arch, and OSId
RunOS='windows'
RunOSId='win'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir\artifacts\bin\mono\$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
xcopy $RunDir\artifacts\bin\runtime\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\shared\Microsoft.NETCore.App\8.0.0\ /e /y
xcopy $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\dotnet-mono\ /e /y
xcopy $RunDir\artifacts\bin\coreclr\$RunOS.$RunArch.Release\corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun /y

# Create Core Root
$RunDir\src\tests\build.cmd release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir\performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir\performance\scripts\benchmarks_ci.py --csproj $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_VectorOf&lt;Single&gt;*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe" 

# Individual Commands:
# Restore 
dotnet restore $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --packages $RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_VectorOf&lt;Single&gt;*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Numerics.Tests.Perf_VectorOf<Single>.OneBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 10/7/2022 12:48:42 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 115.08592327406198 (T) = (0 -1.1414654832546183) / Math.Sqrt((12.126315428938575 / (220)) + (0.3313049015187894 / (3))) is greater than 1.9707562704893595 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (220) + (3) - 2, .975) and 0.9762036471723626 = (47.96808534158684 - 1.1414654832546183) / 47.96808534158684 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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

Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline cd26e63274102c8130109c9ce1301851a550433c
Compare 5d54b08d5fc40d0b1c156f430a487a94c1e34f79
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_Plane

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
CreateFromVector4Benchmark - Duration of single invocation 38.85 ns 25.41 ns 0.65 0.37 False
DotBenchmark - Duration of single invocation 46.88 ns 37.91 ns 0.81 0.30 False

graph
graph
Test Report

Repro

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

Repro Steps

Prerequisites (Build files either built locally or downloaded from payload above)

  • Libraries build extracted to runtime/artifacts or build instructions: Libraries README args: -subset libs+libs.tests -rc release -configuration Release -arch $RunArch -framework net8.0
  • CoreCLR product build extracted to runtime/artifacts/bin/coreclr/$RunOS.$RunArch.Release, build instructions: CoreCLR README args: -subset clr+libs -rc release -configuration Release -arch $RunArch -framework net8.0
  • Mono Runtime build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release
  • Dotnet SDK installed for dotnet commands
  • Running commands from the runtime folder

Linux

# Set $RunDir to the runtime directory
RunDir=`pwd`

# Set the OS, arch, and OSId
RunOS='linux'
RunOSId='linux'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir/artifacts/bin/mono/$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
cp $RunDir/artifacts/bin/runtime/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/shared/Microsoft.NETCore.App/8.0.0 -rf
cp $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/dotnet-mono -r
cp $RunDir/artifacts/bin/coreclr/$RunOS.$RunArch.Release/corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun

# Create Core Root
$RunDir/src/tests/build.sh release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir/performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir/performance/scripts/benchmarks_ci.py --csproj $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Plane*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun" 

# Individual Commands:
# Restore 
dotnet restore $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --packages $RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Plane*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages

Windows

# Set $RunDir to the runtime directory
$RunDir="FullPathHere"

# Set the OS, arch, and OSId
RunOS='windows'
RunOSId='win'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir\artifacts\bin\mono\$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
xcopy $RunDir\artifacts\bin\runtime\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\shared\Microsoft.NETCore.App\8.0.0\ /e /y
xcopy $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\dotnet-mono\ /e /y
xcopy $RunDir\artifacts\bin\coreclr\$RunOS.$RunArch.Release\corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun /y

# Create Core Root
$RunDir\src\tests\build.cmd release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir\performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir\performance\scripts\benchmarks_ci.py --csproj $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Plane*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe" 

# Individual Commands:
# Restore 
dotnet restore $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --packages $RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Plane*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Numerics.Tests.Perf_Plane.CreateFromVector4Benchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 2/2/2023 2:17:49 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 50.52651763328467 (T) = (0 -25.154058944125186) / Math.Sqrt((39.66815486383519 / (220)) + (0.13619426085618158 / (3))) is greater than 1.9707562704893595 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (220) + (3) - 2, .975) and 0.48830767949076825 = (49.15856255003414 - 25.154058944125186) / 49.15856255003414 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Plane.DotBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 2/2/2023 2:17:49 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 17.494718443440053 (T) = (0 -35.4537526474308) / Math.Sqrt((37.87997921993795 / (228)) + (5.1107545166372805 / (3))) is greater than 1.9703772833260038 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (228) + (3) - 2, .975) and 0.40289082117858116 = (59.37566177999446 - 35.4537526474308) / 59.37566177999446 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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

Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline cd26e63274102c8130109c9ce1301851a550433c
Compare 5d54b08d5fc40d0b1c156f430a487a94c1e34f79
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_Quaternion

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
ConjugateBenchmark - Duration of single invocation 57.72 ns 43.69 ns 0.76 0.25 False

graph
Test Report

Repro

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

Repro Steps

Prerequisites (Build files either built locally or downloaded from payload above)

  • Libraries build extracted to runtime/artifacts or build instructions: Libraries README args: -subset libs+libs.tests -rc release -configuration Release -arch $RunArch -framework net8.0
  • CoreCLR product build extracted to runtime/artifacts/bin/coreclr/$RunOS.$RunArch.Release, build instructions: CoreCLR README args: -subset clr+libs -rc release -configuration Release -arch $RunArch -framework net8.0
  • Mono Runtime build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release
  • Dotnet SDK installed for dotnet commands
  • Running commands from the runtime folder

Linux

# Set $RunDir to the runtime directory
RunDir=`pwd`

# Set the OS, arch, and OSId
RunOS='linux'
RunOSId='linux'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir/artifacts/bin/mono/$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
cp $RunDir/artifacts/bin/runtime/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/shared/Microsoft.NETCore.App/8.0.0 -rf
cp $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/dotnet-mono -r
cp $RunDir/artifacts/bin/coreclr/$RunOS.$RunArch.Release/corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun

# Create Core Root
$RunDir/src/tests/build.sh release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir/performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir/performance/scripts/benchmarks_ci.py --csproj $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Quaternion*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun" 

# Individual Commands:
# Restore 
dotnet restore $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --packages $RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Quaternion*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages

Windows

# Set $RunDir to the runtime directory
$RunDir="FullPathHere"

# Set the OS, arch, and OSId
RunOS='windows'
RunOSId='win'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir\artifacts\bin\mono\$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
xcopy $RunDir\artifacts\bin\runtime\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\shared\Microsoft.NETCore.App\8.0.0\ /e /y
xcopy $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\dotnet-mono\ /e /y
xcopy $RunDir\artifacts\bin\coreclr\$RunOS.$RunArch.Release\corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun /y

# Create Core Root
$RunDir\src\tests\build.cmd release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir\performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir\performance\scripts\benchmarks_ci.py --csproj $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Numerics.Tests.Perf_Quaternion*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe" 

# Individual Commands:
# Restore 
dotnet restore $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --packages $RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Numerics.Tests.Perf_Quaternion*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Numerics.Tests.Perf_Quaternion.ConjugateBenchmark

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 2/2/2023 2:17:49 AM, 6/17/2023 12:20:29 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 49.039126780833655 (T) = (0 -44.071732384557784) / Math.Sqrt((15.176878200899427 / (229)) + (0.3124801639029382 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.3147722667132218 = (64.31691282132198 - 44.071732384557784) / 64.31691282132198 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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

@performanceautofiler
Copy link
Author

performanceautofiler bot commented Jul 11, 2023 •
edited
Loading

Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline cd26e63274102c8130109c9ce1301851a550433c
Compare 5d54b08d5fc40d0b1c156f430a487a94c1e34f79
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Linq.Tests.Perf_Enumerable

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
EmptyTakeSelectToArray - Duration of single invocation 187.34 ns 140.91 ns 0.75 0.10 False

graph
Test Report

Repro

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

Repro Steps

Prerequisites (Build files either built locally or downloaded from payload above)

  • Libraries build extracted to runtime/artifacts or build instructions: Libraries README args: -subset libs+libs.tests -rc release -configuration Release -arch $RunArch -framework net8.0
  • CoreCLR product build extracted to runtime/artifacts/bin/coreclr/$RunOS.$RunArch.Release, build instructions: CoreCLR README args: -subset clr+libs -rc release -configuration Release -arch $RunArch -framework net8.0
  • Mono Runtime build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release
  • Dotnet SDK installed for dotnet commands
  • Running commands from the runtime folder

Linux

# Set $RunDir to the runtime directory
RunDir=`pwd`

# Set the OS, arch, and OSId
RunOS='linux'
RunOSId='linux'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir/artifacts/bin/mono/$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
cp $RunDir/artifacts/bin/runtime/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/shared/Microsoft.NETCore.App/8.0.0 -rf
cp $RunDir/artifacts/bin/testhost/net8.0-$RunOS-Release-$RunArch/* $RunDir/artifacts/dotnet-mono -r
cp $RunDir/artifacts/bin/coreclr/$RunOS.$RunArch.Release/corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun

# Create Core Root
$RunDir/src/tests/build.sh release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir/performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir/performance/scripts/benchmarks_ci.py --csproj $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Linq.Tests.Perf_Enumerable*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun" 

# Individual Commands:
# Restore 
dotnet restore $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --packages $RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir/performance/artifacts/packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Linq.Tests.Perf_Enumerable*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir/artifacts/dotnet-mono/shared/Microsoft.NETCore.App/8.0.0/corerun --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages

Windows

# Set $RunDir to the runtime directory
$RunDir="FullPathHere"

# Set the OS, arch, and OSId
RunOS='windows'
RunOSId='win'
RunArch='x64'

# Create mono dotnet
mkdir -p $RunDir/artifacts/dotnet-mono
$RunDir/build.sh -subset libs.pretest -configuration release -ci -arch $RunArch -testscope innerloop /p:RuntimeArtifactsPath=$RunDir\artifacts\bin\mono\$RunOS.$RunArch.Release /p:RuntimeFlavor=mono
xcopy $RunDir\artifacts\bin\runtime\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\shared\Microsoft.NETCore.App\8.0.0\ /e /y
xcopy $RunDir\artifacts\bin\testhost\net8.0-$RunOS-Release-$RunArch\ $RunDir\artifacts\dotnet-mono\ /e /y
xcopy $RunDir\artifacts\bin\coreclr\$RunOS.$RunArch.Release\corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun /y

# Create Core Root
$RunDir\src\tests\build.cmd release $RunArch generatelayoutonly /p:LibrariesConfiguration=Release

# Clone performance 
git clone --branch main --depth 1 --quiet https://github.com/dotnet/performance.git $RunDir\performance

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir\performance\scripts\benchmarks_ci.py --csproj $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Linq.Tests.Perf_Enumerable*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe" 

# Individual Commands:
# Restore 
dotnet restore $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --packages $RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Build
dotnet build $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore /p:NuGetPackageRoot=$RunDir\performance\artifacts\packages /p:UseSharedCompilation=false /p:BuildInParallel=false /m:1

# Run
dotnet run --project $RunDir\performance\src\benchmarks\micro\MicroBenchmarks.csproj --configuration Release --framework net8.0 --no-restore --no-build -- --filter 'System.Linq.Tests.Perf_Enumerable*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Linq.Tests.Perf_Enumerable.EmptyTakeSelectToArray

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because we could not find enough baseline builds for window checking.
IsChangePoint: Marked as a change because one of 3/16/2023 1:27:09 AM, 7/1/2023 1:53:48 AM, 7/5/2023 12:45:20 AM, 7/9/2023 5:37:34 PM falls between 7/1/2023 1:53:48 AM and 7/9/2023 5:37:34 PM.
IsImprovementStdDev: Marked as improvement because 58.578780468982806 (T) = (0 -141.03570446142376) / Math.Sqrt((130.81669560643496 / (229)) + (0.0195391221136835 / (3))) is greater than 1.9703317732749426 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (229) + (3) - 2, .975) and 0.23995347396859104 = (185.56193552760934 - 141.03570446142376) / 185.56193552760934 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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

@kotlarmilos
Copy link
Member

@BrzVlad more improvements by dotnet/runtime#87822.

@kotlarmilos kotlarmilos mentioned this issue Aug 11, 2023
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
ampere arch-arm64 branch-refs/heads/main kind-micro_mono mono-interpreter os-linux perf-improvement PGO Applied if there were any profile guided optimization updates in the observed interval. runtime-mono
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@kotlarmilos