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/x64: 120 Improvements on 4/17/2023 1:30:24 AM #17009

Closed
performanceautofiler bot opened this issue Apr 25, 2023 · 13 comments
Closed

[Perf] Linux/x64: 120 Improvements on 4/17/2023 1:30:24 AM #17009

performanceautofiler bot opened this issue Apr 25, 2023 · 13 comments
Labels
arch-x64 branch-refs/heads/main kind-micro_mono mono-interpreter os-linux perf-improvement runtime-mono

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Text.Json.Tests.Perf_Guids

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
WriteGuids - Duration of single invocation 89.99 ms 68.53 ms 0.76 0.05 True
WriteGuids - Duration of single invocation 86.01 ms 59.01 ms 0.69 0.03 True
WriteGuids - Duration of single invocation 92.84 ms 70.55 ms 0.76 0.04 True
WriteGuids - Duration of single invocation 89.22 ms 60.81 ms 0.68 0.03 True

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Text.Json.Tests.Perf_Guids*' --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.Text.Json.Tests.Perf_Guids*' --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.Text.Json.Tests.Perf_Guids*' --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.Text.Json.Tests.Perf_Guids*' --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

Histogram

System.Text.Json.Tests.Perf_Guids.WriteGuids(Formatted: True, SkipValidation: True)

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 68.53045269642857 < 88.45533994128205.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 75.08007542790556 (T) = (0 -69523193.1901668) / Math.Sqrt((5152692127323.411 / (299)) + (2536491376036.5386 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.24029741265623605 = (91513698.05550984 - 69523193.1901668) / 91513698.05550984 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_Guids.WriteGuids(Formatted: False, SkipValidation: True)

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 59.00672867857143 < 81.87738441732144.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 122.43349652222668 (T) = (0 -59753354.28596179) / Math.Sqrt((3128772686723.1772 / (299)) + (1156061423440.441 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.29500961989720065 = (84757687.4414212 - 59753354.28596179) / 84757687.4414212 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_Guids.WriteGuids(Formatted: True, SkipValidation: False)

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 70.54920676666667 < 90.45248923267856.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 86.81692490288285 (T) = (0 -70402913.36576109) / Math.Sqrt((4745718780057.215 / (299)) + (1834476754551.8718 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.2417264990679234 = (92846332.19968942 - 70402913.36576109) / 92846332.19968942 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_Guids.WriteGuids(Formatted: False, SkipValidation: False)

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 60.813817783333334 < 84.43213395185897.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 132.0954990638115 (T) = (0 -61198563.2016793) / Math.Sqrt((3160604490058.7227 / (299)) + (912665975661.7917 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2883472510726409 = (85994979.00334261 - 61198563.2016793) / 85994979.00334261 is greater than 0.05.
IsChangeEdgeDetector: Marked as 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Text.Json.Tests.Perf_Doubles

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
WriteDoubles - Duration of single invocation 425.03 ms 339.36 ms 0.80 0.02 False
WriteDoubles - Duration of single invocation 402.80 ms 309.92 ms 0.77 0.03 False
WriteDoubles - Duration of single invocation 390.03 ms 305.22 ms 0.78 0.02 False
WriteDoubles - Duration of single invocation 438.59 ms 342.12 ms 0.78 0.03 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Text.Json.Tests.Perf_Doubles*' --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.Text.Json.Tests.Perf_Doubles*' --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.Text.Json.Tests.Perf_Doubles*' --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.Text.Json.Tests.Perf_Doubles*' --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

Histogram

System.Text.Json.Tests.Perf_Doubles.WriteDoubles(Formatted: True, SkipValidation: True)

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 339.3649747142857 < 405.2365703934524.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 58.577779924135115 (T) = (0 -337514712.72922266) / Math.Sqrt((1074394265029481.5 / (299)) + (28527848203428.883 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.26657389236779766 = (460189116.82712984 - 337514712.72922266) / 460189116.82712984 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_Doubles.WriteDoubles(Formatted: False, SkipValidation: False)

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 309.91736180000004 < 381.75292850499994.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 55.048727239173594 (T) = (0 -310952445.42925453) / Math.Sqrt((1175563570471168.8 / (299)) + (39251791918987.3 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.28344196339234373 = (433952910.36211383 - 310952445.42925453) / 433952910.36211383 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_Doubles.WriteDoubles(Formatted: False, SkipValidation: True)

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 305.2241953571429 < 374.96059730416664.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 59.167283237161286 (T) = (0 -304804863.06449866) / Math.Sqrt((1134079254167094.5 / (299)) + (16536435347038.312 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2855612923074277 = (426635426.91986704 - 304804863.06449866) / 426635426.91986704 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_Doubles.WriteDoubles(Formatted: True, SkipValidation: False)

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 342.115343 < 417.5764362232143.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 59.29126980015335 (T) = (0 -344208140.4159934) / Math.Sqrt((1102044990936724.9 / (299)) + (27805027137900.625 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2662434214065131 = (469103992.3291649 - 344208140.4159934) / 469103992.3291649 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

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

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
DivideBenchmark - Duration of single invocation 52.31 ns 43.93 ns 0.84 0.16 False
LessThanAnyBenchmark - Duration of single invocation 22.70 ns 18.74 ns 0.83 0.28 False
LessThanOrEqualAnyBenchmark - Duration of single invocation 20.30 ns 17.96 ns 0.88 0.34 False
DivisionOperatorBenchmark - Duration of single invocation 52.25 ns 43.83 ns 0.84 0.15 True
GreaterThanOrEqualAllBenchmark - Duration of single invocation 19.32 ns 18.31 ns 0.95 0.27 False

graph
graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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;Int64&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;Int64&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;Int64&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;Int64&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

Histogram

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

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 43.932239967672 < 49.38127548058971.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 17.946224376547136 (T) = (0 -44.74838747027099) / Math.Sqrt((2.9167616524478497 / (299)) + (6.920351374799826 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.15103829058079982 = (52.70954740807414 - 44.74838747027099) / 52.70954740807414 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 18.74273588966631 < 19.0833581629203.
IsChangePoint: Marked as a change because one of 4/17/2023 1:54:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 11.593936727910467 (T) = (0 -18.526210494678907) / Math.Sqrt((1.5900704994754322 / (299)) + (0.5142922294400403 / (34))) is greater than 1.9671567996108528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (34) - 2, .975) and 0.08213152453357815 = (20.183948996902494 - 18.526210494678907) / 20.183948996902494 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 17.956276799819452 < 18.626259497615333.
IsChangePoint: Marked as a change because one of 4/19/2023 4:18:46 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.178898933390448 (T) = (0 -18.15558653071902) / Math.Sqrt((2.7262357116058156 / (299)) + (0.8783537910295937 / (26))) is greater than 1.967335607330539 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (26) - 2, .975) and 0.10404233975468954 = (20.263888949560492 - 18.15558653071902) / 20.263888949560492 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 43.830065096816945 < 49.6407563653004.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 18.391951848864135 (T) = (0 -44.745383183379175) / Math.Sqrt((3.3998114059780757 / (299)) + (6.305756571590018 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.14912137941800976 = (52.58726932494073 - 44.745383183379175) / 52.58726932494073 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

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

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 18.31061328278428 < 18.652425474584753.
IsChangePoint: Marked as a change because one of 3/16/2023 12:15:41 AM, 4/18/2023 4:36:42 PM, 4/19/2023 4:18:46 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 15.906365249520288 (T) = (0 -18.24057070678816) / Math.Sqrt((2.617782850725466 / (299)) + (0.2031827323555403 / (26))) is greater than 1.967335607330539 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (26) - 2, .975) and 0.10092268963054038 = (20.288100362907084 - 18.24057070678816) / 20.288100362907084 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_VectorOf<Int16>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.14 ns 2.94 ns 0.71 0.50 False
GreaterThanOrEqualAllBenchmark - Duration of single invocation 19.42 ns 17.98 ns 0.93 0.28 False
InequalityOperatorBenchmark - Duration of single invocation 19.46 ns 15.59 ns 0.80 0.34 False

graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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;Int16&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;Int16&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;Int16&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;Int16&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

Histogram

System.Numerics.Tests.Perf_VectorOf<Int16>.ZeroBenchmark

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 2.941993875564355 < 4.048713668451506.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 9.52426229150513 (T) = (0 -3.007159115126457) / Math.Sqrt((0.4320725927380154 / (299)) + (0.18550529664416632 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.20289209722458235 = (3.772587255321333 - 3.007159115126457) / 3.772587255321333 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<Int16>.GreaterThanOrEqualAllBenchmark

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 17.979762882026897 < 18.759458996289204.
IsChangePoint: Marked as a change because one of 4/17/2023 1:30:24 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 8.399554171730534 (T) = (0 -18.312413351089187) / Math.Sqrt((1.4631251343055507 / (299)) + (2.1155522154876647 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.10494082394011854 = (20.45944429250122 - 18.312413351089187) / 20.45944429250122 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<Int16>.InequalityOperatorBenchmark

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 15.586312001387444 < 19.043826154995266.
IsChangePoint: Marked as a change because one of 4/17/2023 1:30:24 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 8.28055350049648 (T) = (0 -18.163315242031743) / Math.Sqrt((2.2763781328669817 / (299)) + (2.5306807274779506 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.1128860402168657 = (20.474613257660813 - 18.163315242031743) / 20.474613257660813 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Text.Json.Serialization.Tests.WriteJson<Nullable<DateTimeOffset>>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
SerializeToUtf8Bytes - Duration of single invocation 2.82 μs 2.42 μs 0.86 0.12 False
SerializeToString - Duration of single invocation 4.58 μs 4.30 μs 0.94 0.01 False
SerializeToUtf8Bytes - Duration of single invocation 2.91 μs 2.37 μs 0.82 0.13 False
SerializeToWriter - Duration of single invocation 1.43 μs 1.30 μs 0.91 0.04 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Text.Json.Serialization.Tests.WriteJson&lt;Nullable&lt;DateTimeOffset&gt;&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.Text.Json.Serialization.Tests.WriteJson&lt;Nullable&lt;DateTimeOffset&gt;&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.Text.Json.Serialization.Tests.WriteJson&lt;Nullable&lt;DateTimeOffset&gt;&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.Text.Json.Serialization.Tests.WriteJson&lt;Nullable&lt;DateTimeOffset&gt;&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

Histogram

System.Text.Json.Serialization.Tests.WriteJson<Nullable<DateTimeOffset>>.SerializeToUtf8Bytes(Mode: SourceGen)

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 2.4244053510858548 < 2.713183474571993.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 27.29706310105857 (T) = (0 -2587.822779906405) / Math.Sqrt((62888.68579706752 / (299)) + (5814.171373557461 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.1698907893828073 = (3117.4485800276066 - 2587.822779906405) / 3117.4485800276066 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.WriteJson<Nullable<DateTimeOffset>>.SerializeToString(Mode: SourceGen)

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 4.301905443361189 < 4.346240716103336.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 29.453517007717153 (T) = (0 -4290.941184026506) / Math.Sqrt((83470.42598754066 / (299)) + (4014.22526046273 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.11905743003422149 = (4870.852346473838 - 4290.941184026506) / 4870.852346473838 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.WriteJson<Nullable<DateTimeOffset>>.SerializeToUtf8Bytes(Mode: Reflection)

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 2.3697861925596153 < 2.7393584061000706.
IsChangePoint: Marked as a change because one of 3/14/2023 8:42:09 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 24.797138700316864 (T) = (0 -2592.484276391628) / Math.Sqrt((65227.15677863563 / (299)) + (8187.396023114169 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.16900187110030285 = (3119.7233618615587 - 2592.484276391628) / 3119.7233618615587 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.WriteJson<Nullable<DateTimeOffset>>.SerializeToWriter(Mode: SourceGen)

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 1.299693714049775 < 1.3615413255871887.
IsChangePoint: Marked as a change because one of 2/7/2023 12:22:33 AM, 3/16/2023 9:46:36 PM, 3/29/2023 9:08:46 AM, 4/17/2023 1:30:24 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 21.40675112761987 (T) = (0 -1333.034617670749) / Math.Sqrt((1109.062135756751 / (299)) + (479.19035889189485 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.062134218052982505 = (1421.3490281129114 - 1333.034617670749) / 1421.3490281129114 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Tests.Perf_DateTimeOffset

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToString - Duration of single invocation 2.86 μs 2.13 μs 0.74 0.04 False
ToString - Duration of single invocation 2.37 μs 388.32 ns 0.16 0.04 True
ToString - Duration of single invocation 570.50 ns 453.49 ns 0.79 0.05 False

graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Tests.Perf_DateTimeOffset*' --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.Tests.Perf_DateTimeOffset*' --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.Tests.Perf_DateTimeOffset*' --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.Tests.Perf_DateTimeOffset*' --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

Histogram

System.Tests.Perf_DateTimeOffset.ToString(format: "G")

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 2.125699161374931 < 2.753182304123784.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 20.04726163164376 (T) = (0 -2159.680445950601) / Math.Sqrt((18268.961990280688 / (299)) + (3632.706495343099 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10486633116011872 = (2412.690440690951 - 2159.680445950601) / 2412.690440690951 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_DateTimeOffset.ToString(format: "s")

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 388.315226025697 < 2.2606178512452577.
IsChangePoint: Marked as a change because one of 4/19/2023 8:38:27 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 299.46346836065493 (T) = (0 -390.3075344960845) / Math.Sqrt((11593.720700798185 / (299)) + (52.813771138162174 / (23))) is greater than 1.9674049737480528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (23) - 2, .975) and 0.8309972465639831 = (2309.474411278463 - 390.3075344960845) / 2309.474411278463 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_DateTimeOffset.ToString(format: "r")

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 453.48735369477555 < 541.3138867771803.
IsChangePoint: Marked as a change because one of 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 32.677007862029264 (T) = (0 -471.5033293068536) / Math.Sqrt((197.48167486093485 / (299)) + (86.99041285099557 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10735725446686158 = (528.2105653872135 - 471.5033293068536) / 528.2105653872135 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
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 Baseline ETL Compare ETL
AbsBenchmark - Duration of single invocation 17.95 ns 16.62 ns 0.93 0.36 False
CreateFromScalar - Duration of single invocation 12.21 ns 8.07 ns 0.66 0.42 False
UnitZBenchmark - Duration of single invocation 12.53 ns 10.48 ns 0.84 0.36 False
MaxBenchmark - Duration of single invocation 35.72 ns 28.68 ns 0.80 0.21 False
CreateFromScalarXYZWBenchmark - Duration of single invocation 12.04 ns 10.35 ns 0.86 0.38 False
OneBenchmark - Duration of single invocation 9.76 ns 8.20 ns 0.84 0.44 False
MultiplyByScalarBenchmark - Duration of single invocation 30.89 ns 25.28 ns 0.82 0.26 False
EqualityOperatorBenchmark - Duration of single invocation 10.80 ns 9.70 ns 0.90 0.34 False
UnitXBenchmark - Duration of single invocation 12.72 ns 10.33 ns 0.81 0.42 False
MultiplyByScalarOperatorBenchmark - Duration of single invocation 27.96 ns 25.84 ns 0.92 0.18 False

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

Payloads

Baseline
Compare

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

Histogram

System.Numerics.Tests.Perf_Vector4.AbsBenchmark

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 16.624767045768873 < 17.06877101389241.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 7.80762418066988 (T) = (0 -16.325294562978584) / Math.Sqrt((2.183591515226341 / (299)) + (1.0616251428099976 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.08318833467102253 = (17.806595596840072 - 16.325294562978584) / 17.806595596840072 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.CreateFromScalar

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 8.06614311822897 < 9.282790605315565.
IsChangePoint: Marked as a change because one of 3/29/2023 4:47:17 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 7.3769582230077555 (T) = (0 -8.238269777658566) / Math.Sqrt((1.0030090094893829 / (299)) + (0.2841628423450551 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.08673145769976391 = (9.020643322399955 - 8.238269777658566) / 9.020643322399955 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.UnitZBenchmark

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 10.483398605766869 < 11.817105016060523.
IsChangePoint: Marked as a change because one of 2/6/2023 11:23:26 AM, 2/14/2023 9:29:25 PM, 3/28/2023 3:45:48 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 11.591839222835008 (T) = (0 -10.475556509808575) / Math.Sqrt((0.7425021102375585 / (299)) + (0.4512497301785291 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1182190573213435 = (11.879998764756856 - 10.475556509808575) / 11.879998764756856 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.MaxBenchmark

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 28.676729235287116 < 32.784379036393396.
IsChangePoint: Marked as a change because one of 2/15/2023 12:21:17 PM, 3/29/2023 4:47:17 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 9.886564783751442 (T) = (0 -29.49514215218454) / Math.Sqrt((5.389495169923829 / (299)) + (0.8023643764450294 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.06261333954970838 = (31.465288974792944 - 29.49514215218454) / 31.465288974792944 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.CreateFromScalarXYZWBenchmark

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 10.345309032247435 < 11.834603697152811.
IsChangePoint: Marked as a change because one of 2/6/2023 11:23:26 AM, 3/9/2023 1:54:27 PM, 3/29/2023 4:47:17 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 4.490218953728702 (T) = (0 -10.793147320373718) / Math.Sqrt((0.738537922571783 / (299)) + (2.16484158605713 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.09316227050542902 = (11.901961033744499 - 10.793147320373718) / 11.901961033744499 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.OneBenchmark

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 8.196487289265063 < 9.210936817045928.
IsChangePoint: Marked as a change because one of 2/12/2023 3:00:19 AM, 2/12/2023 9:12:37 PM, 3/29/2023 1:05:10 AM, 4/17/2023 7:48:00 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 3.9909789260774704 (T) = (0 -8.442657358649685) / Math.Sqrt((0.7140097084345596 / (299)) + (0.5682443470330183 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.060601229526658355 = (8.98729871063768 - 8.442657358649685) / 8.98729871063768 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.MultiplyByScalarBenchmark

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 25.283621672450643 < 27.276178290718438.
IsChangePoint: Marked as a change because one of 2/6/2023 11:23:26 AM, 2/15/2023 12:21:17 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 8.528066499189709 (T) = (0 -25.888207609773374) / Math.Sqrt((3.6336904029274724 / (299)) + (2.183462024022557 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.08077971741684362 = (28.163224963904582 - 25.888207609773374) / 28.163224963904582 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.EqualityOperatorBenchmark

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 9.697271006804884 < 10.66479724117321.
IsChangePoint: Marked as a change because one of 3/3/2023 1:35:24 AM, 3/3/2023 1:18:24 PM, 4/17/2023 1:30:24 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 11.793629248456465 (T) = (0 -9.978724675690568) / Math.Sqrt((1.086913915962383 / (299)) + (0.12701918578788263 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.09093464804446746 = (10.976905735352108 - 9.978724675690568) / 10.976905735352108 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.UnitXBenchmark

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 10.326872220918903 < 11.858941639879419.
IsChangePoint: Marked as a change because one of 2/6/2023 2:52:08 AM, 2/15/2023 12:21:17 PM, 3/29/2023 4:47:17 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 6.164132148083547 (T) = (0 -10.831579748433718) / Math.Sqrt((1.0203514822887936 / (299)) + (0.9798825899930579 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.09057933798529412 = (11.910417478791091 - 10.831579748433718) / 11.910417478791091 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector4.MultiplyByScalarOperatorBenchmark

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 25.83501161105869 < 27.031520039738737.
IsChangePoint: Marked as a change because one of 2/4/2023 10:21:07 PM, 2/15/2023 12:21:17 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 9.249010269165561 (T) = (0 -25.966816916917207) / Math.Sqrt((1.7659835602356382 / (299)) + (1.492318139592291 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.07114277015712683 = (27.95566001172192 - 25.966816916917207) / 27.95566001172192 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_VectorOf<UInt16>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.53 ns 2.97 ns 0.66 0.53 False
EqualsAllBenchmark - Duration of single invocation 19.25 ns 16.23 ns 0.84 0.25 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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;UInt16&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;UInt16&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;UInt16&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;UInt16&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

Histogram

System.Numerics.Tests.Perf_VectorOf<UInt16>.ZeroBenchmark

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 2.966809621265104 < 3.9768734357151683.
IsChangePoint: Marked as a change because one of 2/15/2023 9:41:15 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 19.879345230671582 (T) = (0 -2.8510392282246437) / Math.Sqrt((0.4991184406121113 / (299)) + (0.02888365047936354 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2565744249199364 = (3.835002889048577 - 2.8510392282246437) / 3.835002889048577 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt16>.EqualsAllBenchmark

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 16.233253134724993 < 18.44294290983096.
IsChangePoint: Marked as a change because one of 3/1/2023 1:48:25 AM, 3/14/2023 5:06:59 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 12.318066907098917 (T) = (0 -18.193767352089992) / Math.Sqrt((3.108649925738366 / (299)) + (0.7197414042525636 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10472349606039248 = (20.321953354108448 - 18.193767352089992) / 20.321953354108448 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in HardwareIntrinsics.RayTracer.SoA

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Render - Duration of single invocation 3.97 ns 2.34 ns 0.59 0.58 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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 'HardwareIntrinsics.RayTracer.SoA*' --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 'HardwareIntrinsics.RayTracer.SoA*' --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 'HardwareIntrinsics.RayTracer.SoA*' --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 'HardwareIntrinsics.RayTracer.SoA*' --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

Histogram

HardwareIntrinsics.RayTracer.SoA.Render

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 2.3352239057065343 < 3.7775211620975355.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 8.365961124001476 (T) = (0 -2.449046657856686) / Math.Sqrt((0.88588424739803 / (299)) + (0.21203558521013477 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2415653563369942 = (3.229080683905135 - 2.449046657856686) / 3.229080683905135 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_VectorOf<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.46 ns 2.86 ns 0.64 0.55 False
GreaterThanAllBenchmark - Duration of single invocation 22.31 ns 17.98 ns 0.81 0.23 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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;Int32&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;Int32&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;Int32&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;Int32&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

Histogram

System.Numerics.Tests.Perf_VectorOf<Int32>.ZeroBenchmark

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 2.8569190592536673 < 4.2393826207085565.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 13.955589919199213 (T) = (0 -2.9411601774405445) / Math.Sqrt((0.6400352610908683 / (299)) + (0.0720844974064378 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2327810477830474 = (3.833534311087833 - 2.9411601774405445) / 3.833534311087833 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<Int32>.GreaterThanAllBenchmark

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 17.98130730807189 < 19.027673575833994.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 6.399787675712037 (T) = (0 -18.68182118671879) / Math.Sqrt((2.5210107843320695 / (299)) + (1.9265690472415824 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.07771295268628234 = (20.255972629272037 - 18.68182118671879) / 20.255972629272037 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
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 Baseline ETL Compare ETL
MultiplyByScalarOperatorBenchmark - Duration of single invocation 26.11 ns 23.33 ns 0.89 0.25 False
NegateBenchmark - Duration of single invocation 30.36 ns 26.00 ns 0.86 0.25 False
AddOperatorBenchmark - Duration of single invocation 36.68 ns 32.90 ns 0.90 0.21 False
EqualityOperatorBenchmark - Duration of single invocation 34.34 ns 30.78 ns 0.90 0.25 False
CreateFromScalarXYZWBenchmark - Duration of single invocation 12.75 ns 10.45 ns 0.82 0.32 False
IdentityBenchmark - Duration of single invocation 12.61 ns 10.47 ns 0.83 0.34 False

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

Payloads

Baseline
Compare

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

Histogram

System.Numerics.Tests.Perf_Quaternion.MultiplyByScalarOperatorBenchmark

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 23.334305307631794 < 24.347773053926392.
IsChangePoint: Marked as a change because one of 2/8/2023 7:09:09 PM, 2/15/2023 12:21:17 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 6.713532322155714 (T) = (0 -23.49539440243773) / Math.Sqrt((1.7016282090715185 / (299)) + (1.8219567517452346 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.0634945716532539 = (25.088369689340908 - 23.49539440243773) / 25.088369689340908 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Quaternion.NegateBenchmark

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 25.99609143478481 < 27.4190523968939.
IsChangePoint: Marked as a change because one of 3/6/2023 11:14:52 AM, 3/6/2023 8:17:12 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 9.711586997142085 (T) = (0 -26.628166082870013) / Math.Sqrt((4.840936940721742 / (299)) + (1.5853375952542457 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.08140250451707941 = (28.987849644496563 - 26.628166082870013) / 28.987849644496563 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Quaternion.AddOperatorBenchmark

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 32.89509491595129 < 34.846685130891316.
IsChangePoint: Marked as a change because one of 2/6/2023 4:11:28 AM, 2/15/2023 12:21:17 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.192358802629188 (T) = (0 -33.956904023201346) / Math.Sqrt((3.9455703045851425 / (299)) + (2.4615942759044356 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.07812950208170824 = (36.83478764086781 - 33.956904023201346) / 36.83478764086781 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Quaternion.EqualityOperatorBenchmark

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 30.775515614138158 < 32.4892102170192.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 9.244366946500634 (T) = (0 -31.50432582136217) / Math.Sqrt((3.15570643439347 / (299)) + (3.5375119095076517 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.08726452974305716 = (34.5163816329976 - 31.50432582136217) / 34.5163816329976 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Quaternion.CreateFromScalarXYZWBenchmark

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 10.448803834556953 < 11.884531490949865.
IsChangePoint: Marked as a change because one of 3/10/2023 4:14:40 PM, 3/29/2023 4:47:17 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 4.676065186517042 (T) = (0 -10.717718055449506) / Math.Sqrt((0.6269977565665701 / (299)) + (2.0690240740844508 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.09509489918140139 = (11.844024357641485 - 10.717718055449506) / 11.844024357641485 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Quaternion.IdentityBenchmark

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 10.467909243510801 < 11.872427578123842.
IsChangePoint: Marked as a change because one of 2/6/2023 11:23:26 AM, 3/11/2023 4:36:40 PM, 3/29/2023 4:47:17 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 8.88054506280757 (T) = (0 -10.682563530027767) / Math.Sqrt((1.0168701253204646 / (299)) + (0.600385749241474 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10431622006658206 = (11.926713165244404 - 10.682563530027767) / 11.926713165244404 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Tests.Perf_DateTime

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToString - Duration of single invocation 427.17 ns 325.97 ns 0.76 0.07 False
ToString - Duration of single invocation 2.22 μs 287.67 ns 0.13 0.03 True
ToString - Duration of single invocation 2.72 μs 1.97 μs 0.72 0.03 False
ToString - Duration of single invocation 2.68 μs 1.99 μs 0.74 0.04 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Tests.Perf_DateTime*' --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.Tests.Perf_DateTime*' --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.Tests.Perf_DateTime*' --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.Tests.Perf_DateTime*' --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

Histogram

System.Tests.Perf_DateTime.ToString(format: "r")

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 325.96631926373675 < 403.80538305383465.
IsChangePoint: Marked as a change because one of 3/17/2023 8:17:18 AM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 50.11226311131236 (T) = (0 -333.6840416508361) / Math.Sqrt((254.02337251082912 / (299)) + (47.27141184771959 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1796781291937328 = (406.77208974432085 - 333.6840416508361) / 406.77208974432085 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_DateTime.ToString(format: "s")

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 287.6663280284946 < 2.1250929837467765.
IsChangePoint: Marked as a change because one of 4/19/2023 8:38:27 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 290.4374078311032 (T) = (0 -287.13946701777337) / Math.Sqrt((12065.842925722633 / (299)) + (54.23181734153309 / (23))) is greater than 1.9674049737480528 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (23) - 2, .975) and 0.8686025285339098 = (2185.273915958692 - 287.13946701777337) / 2185.273915958692 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_DateTime.ToString(format: "G")

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 1.9715892256909122 < 2.570887240694435.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 21.44809464984385 (T) = (0 -2008.1270757756158) / Math.Sqrt((15145.122291591446 / (299)) + (3260.9423692979713 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1117633235400076 = (2260.8017986589693 - 2008.1270757756158) / 2260.8017986589693 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_DateTime.ToString(format: null)

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 1.9915493091040708 < 2.557743894169403.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 22.298709898069724 (T) = (0 -2002.650059597586) / Math.Sqrt((15578.384409526685 / (299)) + (3406.2702492722087 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.11792528042774879 = (2270.3859606913356 - 2002.650059597586) / 2270.3859606913356 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.54 ns 3.13 ns 0.69 0.54 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt16&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt16&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt16&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt16&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>.ZeroBenchmark

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 3.125753725533058 < 4.126634174209639.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.134999181838213 (T) = (0 -2.7744626513677244) / Math.Sqrt((0.3512708299503345 / (299)) + (0.29361903595948274 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2585290409576887 = (3.7418358973239334 - 2.7744626513677244) / 3.7418358973239334 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Memory.Span<Char>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToArray - Duration of single invocation 98.30 ns 90.94 ns 0.93 0.09 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Memory.Span&lt;Char&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.Memory.Span&lt;Char&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.Memory.Span&lt;Char&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.Memory.Span&lt;Char&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

Histogram

System.Memory.Span<Char>.ToArray(Size: 33)

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 90.94023209558374 < 93.22065515268123.
IsChangePoint: Marked as a change because one of 3/21/2023 12:39:01 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 29.19716436562789 (T) = (0 -89.21531514400371) / Math.Sqrt((17.30181124683407 / (299)) + (4.360621288977416 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.12063694828815218 = (101.45447317843193 - 89.21531514400371) / 101.45447317843193 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Byte>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.31 ns 2.89 ns 0.67 0.53 False
GreaterThanAnyBenchmark - Duration of single invocation 25.41 ns 23.76 ns 0.94 0.26 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Byte&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Byte&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Byte&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Byte&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Byte>.ZeroBenchmark

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 2.891800050031482 < 4.167746567324042.
IsChangePoint: Marked as a change because one of 2/15/2023 8:55:14 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.114934367978652 (T) = (0 -2.7884754272474357) / Math.Sqrt((0.45606990177597595 / (299)) + (0.3204029991701429 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.26797742485335735 = (3.809275180739929 - 2.7884754272474357) / 3.809275180739929 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Byte>.GreaterThanAnyBenchmark

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 23.762393940409808 < 24.615375421116983.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 6.180621606206648 (T) = (0 -24.238755635470646) / Math.Sqrt((1.357712216822059 / (299)) + (1.975504349939587 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.05782502867439039 = (25.72638456036197 - 24.238755635470646) / 25.72638456036197 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Buffers.Text.Tests.Utf8FormatterTests

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
FormatterDouble - Duration of single invocation 3.57 μs 2.70 μs 0.76 0.05 False
FormatterDouble - Duration of single invocation 3.43 μs 2.62 μs 0.76 0.05 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Buffers.Text.Tests.Utf8FormatterTests*' --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.Buffers.Text.Tests.Utf8FormatterTests*' --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.Buffers.Text.Tests.Utf8FormatterTests*' --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.Buffers.Text.Tests.Utf8FormatterTests*' --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

Histogram

System.Buffers.Text.Tests.Utf8FormatterTests.FormatterDouble(value: -1.7976931348623157E+308)

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 2.703416968785796 < 3.4206471049115263.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 48.43341989005725 (T) = (0 -2745.705937407276) / Math.Sqrt((117068.40569659803 / (299)) + (5927.096626289396 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2929543121339453 = (3883.350092543712 - 2745.705937407276) / 3883.350092543712 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Buffers.Text.Tests.Utf8FormatterTests.FormatterDouble(value: 1.7976931348623157E+308)

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 2.6196927223421325 < 3.4462698496088455.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 57.562324871948626 (T) = (0 -2651.304255144761) / Math.Sqrt((114188.4538106316 / (299)) + (3940.85478306509 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.32423862004440684 = (3923.432640259774 - 2651.304255144761) / 3923.432640259774 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.MathBenchmarks.MathTests

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
DivRemInt64 - Duration of single invocation 14.98 ns 10.71 ns 0.71 0.26 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.MathBenchmarks.MathTests*' --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.MathBenchmarks.MathTests*' --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.MathBenchmarks.MathTests*' --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.MathBenchmarks.MathTests*' --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

Histogram

System.MathBenchmarks.MathTests.DivRemInt64

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 10.7054627956478 < 14.200495235151148.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 25.964969904037137 (T) = (0 -10.97537424297687) / Math.Sqrt((0.9782571888471506 / (299)) + (0.9276814975900236 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.284841455247591 = (15.346770759449699 - 10.97537424297687) / 15.346770759449699 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int64>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.44 ns 3.07 ns 0.69 0.59 False
DotBenchmark - Duration of single invocation 42.31 ns 31.31 ns 0.74 0.21 False
GreaterThanOrEqualAnyBenchmark - Duration of single invocation 26.57 ns 23.89 ns 0.90 0.21 False

graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int64&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int64&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int64&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int64&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int64>.ZeroBenchmark

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 3.068071018506999 < 4.212851287598924.
IsChangePoint: Marked as a change because one of 2/17/2023 10:37:11 AM, 3/14/2023 8:42:09 PM, 4/5/2023 11:58:06 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 15.838476661817007 (T) = (0 -2.964822089880333) / Math.Sqrt((0.4824320892066097 / (299)) + (0.05274461478159395 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.22749931496011172 = (3.837953994470883 - 2.964822089880333) / 3.837953994470883 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int64>.DotBenchmark

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 31.311867731111267 < 42.549828522683455.
IsChangePoint: Marked as a change because one of 2/6/2023 11:23:26 AM, 3/14/2023 5:06:59 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 18.553395769378703 (T) = (0 -30.392452993249112) / Math.Sqrt((48.685908563154975 / (299)) + (0.46479715912361325 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.20393696187741484 = (38.178450119887366 - 30.392452993249112) / 38.178450119887366 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int64>.GreaterThanOrEqualAnyBenchmark

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 23.886996302789612 < 25.1085452629253.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 16.240625146369958 (T) = (0 -24.47700514508157) / Math.Sqrt((2.2358608240066857 / (299)) + (0.11316161554627817 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.0637635338849471 = (26.14404162940778 - 24.47700514508157) / 26.14404162940778 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
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 Baseline ETL Compare ETL
EqualityOperatorBenchmark - Duration of single invocation 28.85 ns 21.24 ns 0.74 0.25 False
LessThanAnyBenchmark - Duration of single invocation 22.66 ns 21.51 ns 0.95 0.25 False
ZeroBenchmark - Duration of single invocation 4.32 ns 2.62 ns 0.61 0.57 False
InequalityOperatorBenchmark - Duration of single invocation 22.49 ns 20.62 ns 0.92 0.27 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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

Histogram

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

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 21.23508847593291 < 21.492618195320883.
IsChangePoint: Marked as a change because one of 3/16/2023 10:20:48 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 16.301989178011343 (T) = (0 -21.15728016689618) / Math.Sqrt((1.836820694842864 / (299)) + (0.4138029832336378 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.09283049737710627 = (23.322300965502325 - 21.15728016689618) / 23.322300965502325 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 21.5074505341162 < 21.754106495797117.
IsChangePoint: Marked as a change because one of 3/18/2023 12:49:26 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.037441681170307 (T) = (0 -21.318001474427604) / Math.Sqrt((1.7185884907486748 / (299)) + (1.3882708141601996 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.09017563852178946 = (23.430897629287266 - 21.318001474427604) / 23.430897629287266 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 2.622169723279266 < 4.07758367358776.
IsChangePoint: Marked as a change because one of 4/3/2023 9:14:01 PM, 4/4/2023 11:36:00 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 16.104595809827416 (T) = (0 -2.8017737454702742) / Math.Sqrt((0.7715184730235597 / (299)) + (0.06603678538254194 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.27523857532266255 = (3.865787623448115 - 2.8017737454702742) / 3.865787623448115 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 20.618163122961356 < 21.28082266736697.
IsChangePoint: Marked as a change because one of 3/19/2023 10:05:03 PM, 3/21/2023 3:33:39 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 7.100834699595535 (T) = (0 -21.524758775471575) / Math.Sqrt((2.3319474752226435 / (299)) + (2.2714131879951505 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.07984533216761203 = (23.392544240608522 - 21.524758775471575) / 23.392544240608522 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_Vector2

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ClampBenchmark - Duration of single invocation 43.45 ns 37.28 ns 0.86 0.19 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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_Vector2*' --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_Vector2*' --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_Vector2*' --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_Vector2*' --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

Histogram

System.Numerics.Tests.Perf_Vector2.ClampBenchmark

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 37.28228684947202 < 42.017878696586585.
IsChangePoint: Marked as a change because one of 2/17/2023 12:28:21 PM, 3/6/2023 4:48:38 PM, 3/29/2023 1:05:10 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 7.74532173717243 (T) = (0 -38.27045918917924) / Math.Sqrt((6.529794763111801 / (299)) + (2.867226974507834 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.059960490113579734 = (40.711543277370595 - 38.27045918917924) / 40.711543277370595 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Memory.Span<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Reverse - Duration of single invocation 1.86 μs 1.69 μs 0.91 0.08 True
Fill - Duration of single invocation 154.12 ns 116.50 ns 0.76 0.07 True
ToArray - Duration of single invocation 107.10 ns 98.79 ns 0.92 0.08 False
Fill - Duration of single invocation 1.97 μs 1.34 μs 0.68 0.07 True

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Memory.Span&lt;Int32&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.Memory.Span&lt;Int32&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.Memory.Span&lt;Int32&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.Memory.Span&lt;Int32&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

Histogram

System.Memory.Span<Int32>.Reverse(Size: 512)

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 1.6904008781216677 < 1.7718828655816903.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 17.788351203814045 (T) = (0 -1567.894605266858) / Math.Sqrt((3437.0675043972938 / (299)) + (10641.71073746421 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.16403031155989814 = (1875.5400189120605 - 1567.894605266858) / 1875.5400189120605 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Memory.Span<Int32>.Fill(Size: 33)

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 116.50486969501479 < 148.33509242026955.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 74.69022474059517 (T) = (0 -116.41063595739881) / Math.Sqrt((7.723524422059529 / (299)) + (9.622536498431218 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2554350862436472 = (156.34719526347754 - 116.41063595739881) / 156.34719526347754 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Memory.Span<Int32>.ToArray(Size: 33)

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 98.78622263511187 < 101.85293464868936.
IsChangePoint: Marked as a change because one of 3/21/2023 12:39:01 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 25.05802138081613 (T) = (0 -99.68472093831286) / Math.Sqrt((17.007651871361592 / (299)) + (6.022939967068295 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10539692694712835 = (111.42899453512321 - 99.68472093831286) / 111.42899453512321 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Memory.Span<Int32>.Fill(Size: 512)

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 1.342686376970038 < 1.8536013692616955.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 127.96532905105451 (T) = (0 -1342.2821067147895) / Math.Sqrt((3186.6948721988856 / (299)) + (477.02319489807104 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.3196469042730663 = (1972.9198193485217 - 1342.2821067147895) / 1972.9198193485217 is greater than 0.05.
IsChangeEdgeDetector: Marked as 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt64>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 5.20 ns 2.97 ns 0.57 0.50 False
DotBenchmark - Duration of single invocation 42.56 ns 29.50 ns 0.69 0.17 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt64&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt64&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt64&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt64&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt64>.ZeroBenchmark

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 2.9651283076528143 < 4.229603558500905.
IsChangePoint: Marked as a change because one of 2/16/2023 1:00:08 AM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 17.118873915995675 (T) = (0 -2.9027387531779967) / Math.Sqrt((0.4117312901898071 / (299)) + (0.04260025510665447 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.22871872286841127 = (3.763528091817998 - 2.9027387531779967) / 3.763528091817998 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt64>.DotBenchmark

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 29.504715802300137 < 42.24971905066289.
IsChangePoint: Marked as a change because one of 2/6/2023 11:23:26 AM, 3/14/2023 5:06:59 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 16.990939963275466 (T) = (0 -30.43791058932598) / Math.Sqrt((48.74389321389247 / (299)) + (1.5024051713446147 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.20212394107990803 = (38.148670146241805 - 30.43791058932598) / 38.148670146241805 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_VectorOf<UInt32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
GreaterThanOrEqualAnyBenchmark - Duration of single invocation 53.43 ns 49.96 ns 0.93 0.18 False
GreaterThanOrEqualAllBenchmark - Duration of single invocation 19.47 ns 18.22 ns 0.94 0.33 False
LessThanBenchmark - Duration of single invocation 76.86 ns 69.97 ns 0.91 0.16 False
ZeroBenchmark - Duration of single invocation 7.12 ns 2.95 ns 0.41 0.53 False
GreaterThanBenchmark - Duration of single invocation 74.64 ns 70.17 ns 0.94 0.10 False
LessThanAnyBenchmark - Duration of single invocation 19.31 ns 18.25 ns 0.95 0.27 False

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

Payloads

Baseline
Compare

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;UInt32&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;UInt32&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;UInt32&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;UInt32&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

Histogram

System.Numerics.Tests.Perf_VectorOf<UInt32>.GreaterThanOrEqualAnyBenchmark

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 49.95935104971123 < 50.456914735491615.
IsChangePoint: Marked as a change because one of 2/28/2023 7:53:18 AM, 3/8/2023 2:17:54 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 6.144015787738812 (T) = (0 -50.72609497479435) / Math.Sqrt((5.909648981889287 / (299)) + (13.62838827367283 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.07101524078227715 = (54.60379674851679 - 50.72609497479435) / 54.60379674851679 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt32>.GreaterThanOrEqualAllBenchmark

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 18.223942703835863 < 19.369511789667026.
IsChangePoint: Marked as a change because one of 3/14/2023 2:33:56 PM, 4/19/2023 1:39:34 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 9.526377642015095 (T) = (0 -18.113957509287008) / Math.Sqrt((1.9435770934940733 / (299)) + (1.1555610529941593 / (27))) is greater than 1.9673127716992511 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (27) - 2, .975) and 0.1045607775938251 = (20.22913119732703 - 18.113957509287008) / 20.22913119732703 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt32>.LessThanBenchmark

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 69.96597237053727 < 69.97076275821735.
IsChangePoint: Marked as a change because one of 2/6/2023 11:23:26 AM, 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 3/17/2023 10:06:19 AM, 4/18/2023 4:36:42 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 6.399100007977095 (T) = (0 -70.67696163325995) / Math.Sqrt((6.00403026049914 / (299)) + (11.869176583032383 / (29))) is greater than 1.9672675222597202 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (29) - 2, .975) and 0.05600449885779912 = (74.87001955808407 - 70.67696163325995) / 74.87001955808407 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt32>.ZeroBenchmark

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 2.9508125856314487 < 4.123195243008337.
IsChangePoint: Marked as a change because one of 2/15/2023 8:55:14 PM, 3/14/2023 5:06:59 PM, 4/6/2023 9:09:21 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 15.843087409898637 (T) = (0 -2.7874705843305256) / Math.Sqrt((0.5479583845605358 / (299)) + (0.08178748617841027 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.2669337470896397 = (3.8024811171758834 - 2.7874705843305256) / 3.8024811171758834 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt32>.GreaterThanBenchmark

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 70.17124939622519 < 71.09096078971942.
IsChangePoint: Marked as a change because one of 2/6/2023 9:02:07 PM, 2/15/2023 6:58:43 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 26.80705258943008 (T) = (0 -70.28186540298609) / Math.Sqrt((4.707618634440894 / (299)) + (0.7955399084080308 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.06856404093533773 = (75.45539198804646 - 70.28186540298609) / 75.45539198804646 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<UInt32>.LessThanAnyBenchmark

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 18.25419943312863 < 18.656168858242193.
IsChangePoint: Marked as a change because one of 3/15/2023 5:33:41 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 15.332591238181449 (T) = (0 -18.32001330929462) / Math.Sqrt((2.098540402917818 / (299)) + (0.3948946500441992 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.10092004223315462 = (20.376400509247556 - 18.32001330929462) / 20.376400509247556 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_BigInteger

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ModPow - Duration of single invocation 353.33 ns 279.80 ns 0.79 0.09 True

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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_BigInteger*' --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_BigInteger*' --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_BigInteger*' --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_BigInteger*' --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

Histogram

System.Numerics.Tests.Perf_BigInteger.ModPow(arguments: 16,16,16 bits)

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 279.7965372547885 < 335.37987239399837.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 38.27368843544152 (T) = (0 -284.4149848783012) / Math.Sqrt((147.57819681363767 / (299)) + (136.94524337135107 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.21606455736402105 = (362.80408999235607 - 284.4149848783012) / 362.80408999235607 is greater than 0.05.
IsChangeEdgeDetector: Marked as 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
DeserializeFromReader - Duration of single invocation 969.45 μs 848.75 μs 0.88 0.00 True
DeserializeFromString - Duration of single invocation 746.32 μs 621.97 μs 0.83 0.00 True
DeserializeFromStream - Duration of single invocation 752.15 μs 622.31 μs 0.83 0.00 True
DeserializeFromUtf8Bytes - Duration of single invocation 870.37 μs 746.49 μs 0.86 0.00 True
DeserializeFromReader - Duration of single invocation 844.64 μs 716.64 μs 0.85 0.01 True
DeserializeFromUtf8Bytes - Duration of single invocation 739.67 μs 613.47 μs 0.83 0.00 True
DeserializeFromStream - Duration of single invocation 873.91 μs 746.62 μs 0.85 0.00 True
DeserializeFromString - Duration of single invocation 869.82 μs 745.81 μs 0.86 0.00 True

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

Payloads

Baseline
Compare

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.Text.Json.Serialization.Tests.ReadJson&lt;TreeRecord&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.Text.Json.Serialization.Tests.ReadJson&lt;TreeRecord&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.Text.Json.Serialization.Tests.ReadJson&lt;TreeRecord&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.Text.Json.Serialization.Tests.ReadJson&lt;TreeRecord&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

Histogram

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromReader(Mode: SourceGen)

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 848.7486630819838 < 928.052049937221.
IsChangePoint: Marked as a change because one of 2/28/2023 7:53:18 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 84.31787968670488 (T) = (0 -845913.8809788389) / Math.Sqrt((342317041.6777152 / (299)) + (44888930.10333758 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1327455600128407 = (975392.9665570391 - 845913.8809788389) / 975392.9665570391 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromString(Mode: Reflection)

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 621.9714148214284 < 713.7073273859127.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 123.91370298174314 (T) = (0 -617037.0231884528) / Math.Sqrt((135539083.90168583 / (299)) + (25424171.24103445 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.17779856946237935 = (750469.4108656378 - 617037.0231884528) / 750469.4108656378 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromStream(Mode: Reflection)

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 622.3130115327381 < 714.8136894086593.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 103.66965428487953 (T) = (0 -618018.8801848138) / Math.Sqrt((174447521.32754934 / (299)) + (39644705.70566154 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.17749170049326074 = (751383.1538908988 - 618018.8801848138) / 751383.1538908988 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromUtf8Bytes(Mode: SourceGen)

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 746.4873667582418 < 822.643536142113.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 91.13630754608845 (T) = (0 -737671.5364214167) / Math.Sqrt((237416000.83491594 / (299)) + (45313495.239200436 / (35))) is greater than 1.9671350567190904 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (35) - 2, .975) and 0.1515018181727845 = (869384.9347241535 - 737671.5364214167) / 869384.9347241535 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromReader(Mode: Reflection)

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 716.6388137175323 < 810.7902037946427.
IsChangePoint: Marked as a change because one of 2/28/2023 7:53:18 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 84.01370014628583 (T) = (0 -716591.4550759498) / Math.Sqrt((363807434.8991563 / (299)) + (41608528.08952988 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.1529633645925032 = (845998.1836927375 - 716591.4550759498) / 845998.1836927375 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromUtf8Bytes(Mode: Reflection)

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 613.4687781593407 < 702.7908069155844.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 104.22288614264033 (T) = (0 -608648.6988370155) / Math.Sqrt((155093454.39809826 / (299)) + (38146125.06197368 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.17704050311904454 = (739585.2422188636 - 608648.6988370155) / 739585.2422188636 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromStream(Mode: SourceGen)

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 746.6169605654762 < 830.2525939651059.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 103.27831706516068 (T) = (0 -746480.438924968) / Math.Sqrt((259637028.6882288 / (299)) + (27672745.818307046 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1495819668396783 = (877780.5853327204 - 746480.438924968) / 877780.5853327204 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Serialization.Tests.ReadJson<TreeRecord>.DeserializeFromString(Mode: SourceGen)

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 745.8130262896825 < 826.0731705605159.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 100.0834307867649 (T) = (0 -746925.6650937396) / Math.Sqrt((230670548.35218832 / (299)) + (34829047.77150192 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1491990246119527 = (877908.8020592236 - 746925.6650937396) / 877908.8020592236 is greater than 0.05.
IsChangeEdgeDetector: Marked as 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Memory.Constructors_ValueTypesOnly<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ReadOnlyFromPointerLength - Duration of single invocation 8.24 ns 6.62 ns 0.80 0.32 False
SpanFromPointerLength - Duration of single invocation 8.09 ns 7.04 ns 0.87 0.40 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Memory.Constructors_ValueTypesOnly&lt;Int32&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.Memory.Constructors_ValueTypesOnly&lt;Int32&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.Memory.Constructors_ValueTypesOnly&lt;Int32&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.Memory.Constructors_ValueTypesOnly&lt;Int32&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

Histogram

System.Memory.Constructors_ValueTypesOnly<Int32>.ReadOnlyFromPointerLength

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 6.624284233239028 < 7.593833057960984.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 7.352041224797811 (T) = (0 -7.054500162708911) / Math.Sqrt((0.46226959238611304 / (299)) + (0.17389465887709063 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.07609672667466055 = (7.635539743590419 - 7.054500162708911) / 7.635539743590419 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Memory.Constructors_ValueTypesOnly<Int32>.SpanFromPointerLength

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 7.040718547361133 < 7.624354072625682.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 4.3976004608775225 (T) = (0 -7.149769556809783) / Math.Sqrt((0.5328729753655828 / (299)) + (0.48607498712687275 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.06987771023424125 = (7.686913468776644 - 7.149769556809783) / 7.686913468776644 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Memory.ReadOnlySpan

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
StringAsSpan - Duration of single invocation 16.38 ns 14.23 ns 0.87 0.29 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Memory.ReadOnlySpan*' --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.Memory.ReadOnlySpan*' --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.Memory.ReadOnlySpan*' --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.Memory.ReadOnlySpan*' --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

Histogram

System.Memory.ReadOnlySpan.StringAsSpan

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 14.228052791057367 < 15.004366746117398.
IsChangePoint: Marked as a change because one of 3/28/2023 4:36:21 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 7.459167436221449 (T) = (0 -14.579066418651653) / Math.Sqrt((1.2305052192250125 / (299)) + (0.5518519420341937 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.06592716808091473 = (15.608061727583332 - 14.579066418651653) / 15.608061727583332 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Float

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.27 ns 2.94 ns 0.69 0.59 False
InequalityOperatorBenchmark - Duration of single invocation 29.15 ns 26.57 ns 0.91 0.28 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Float*' --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.Runtime.Intrinsics.Tests.Perf_Vector128Float*' --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.Runtime.Intrinsics.Tests.Perf_Vector128Float*' --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.Runtime.Intrinsics.Tests.Perf_Vector128Float*' --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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Float.ZeroBenchmark

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 2.9419512961691057 < 4.05314430206651.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 2.562850628355629 (T) = (0 -3.180929554694919) / Math.Sqrt((0.6282988269485524 / (299)) + (2.2115050106910292 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.16695001670221235 = (3.818413802858024 - 3.180929554694919) / 3.818413802858024 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Float.InequalityOperatorBenchmark

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 26.574468985874084 < 27.089646596901986.
IsChangePoint: Marked as a change because one of 4/20/2023 12:14:19 AM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 12.592549123168405 (T) = (0 -26.88444541504559) / Math.Sqrt((4.408751509069127 / (299)) + (0.23955123294174524 / (21))) is greater than 1.967451947860886 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (21) - 2, .975) and 0.07041357205732976 = (28.920866965049555 - 26.88444541504559) / 28.920866965049555 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

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

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsAnyBenchmark - Duration of single invocation 32.57 ns 30.43 ns 0.93 0.23 False
LessThanAnyBenchmark - Duration of single invocation 19.83 ns 18.04 ns 0.91 0.28 False
EqualityOperatorBenchmark - Duration of single invocation 22.23 ns 17.61 ns 0.79 0.22 False
DivideBenchmark - Duration of single invocation 52.17 ns 45.18 ns 0.87 0.13 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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;UInt64&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;UInt64&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;UInt64&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;UInt64&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

Histogram

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

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 30.42622020136902 < 30.831369334147844.
IsChangePoint: Marked as a change because one of 2/28/2023 7:53:18 AM, 3/14/2023 5:06:59 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 6.891875282539709 (T) = (0 -31.014354572148374) / Math.Sqrt((2.919992565320818 / (299)) + (3.131046921300214 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.06390773680832543 = (33.13172834737751 - 31.014354572148374) / 33.13172834737751 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 18.044468590292272 < 18.838257647187287.
IsChangePoint: Marked as a change because one of 2/28/2023 11:10:36 AM, 3/14/2023 2:33:56 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 3.361402677112994 (T) = (0 -18.907444080183506) / Math.Sqrt((1.2622384813846663 / (299)) + (6.0171428381752206 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.06770202795716788 = (20.280473246931884 - 18.907444080183506) / 20.280473246931884 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 17.606296409651062 < 18.707798215271154.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 18.7400188233671 (T) = (0 -18.12661444446789) / Math.Sqrt((1.5205501035815785 / (299)) + (0.2949381635053112 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1056519448555175 = (20.267964289964855 - 18.12661444446789) / 20.267964289964855 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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

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 45.18358124995948 < 46.60973617531429.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 20.586007545259527 (T) = (0 -44.85942328972703) / Math.Sqrt((2.403246648401996 / (299)) + (1.5381959404023462 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.09273415261871376 = (49.44462906788387 - 44.85942328972703) / 49.44462906788387 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Collections.CtorGivenSize<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
List - Duration of single invocation 241.92 ns 210.84 ns 0.87 0.35 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Collections.CtorGivenSize&lt;Int32&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.Collections.CtorGivenSize&lt;Int32&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.Collections.CtorGivenSize&lt;Int32&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.Collections.CtorGivenSize&lt;Int32&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

Histogram

System.Collections.CtorGivenSize<Int32>.List(Size: 512)

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 210.83784379407618 < 230.33491386892402.
IsChangePoint: Marked as a change because one of 2/14/2023 11:21:41 AM, 2/28/2023 7:53:18 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.806153031988183 (T) = (0 -217.3661949771758) / Math.Sqrt((164.26526915306982 / (299)) + (85.46194005691977 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.07834080291300105 = (235.8422675802342 - 217.3661949771758) / 235.8422675802342 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Text.Json.Tests.Perf_DateTimes

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
WriteDateTimes - Duration of single invocation 66.07 ms 58.02 ms 0.88 0.07 False
WriteDateTimes - Duration of single invocation 70.96 ms 63.08 ms 0.89 0.06 False
WriteDateTimes - Duration of single invocation 63.70 ms 58.80 ms 0.92 0.05 False
WriteDateTimes - Duration of single invocation 74.36 ms 66.68 ms 0.90 0.06 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Text.Json.Tests.Perf_DateTimes*' --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.Text.Json.Tests.Perf_DateTimes*' --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.Text.Json.Tests.Perf_DateTimes*' --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.Text.Json.Tests.Perf_DateTimes*' --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

Histogram

System.Text.Json.Tests.Perf_DateTimes.WriteDateTimes(Formatted: False, SkipValidation: False)

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 58.02017158333334 < 62.789612882499995.
IsChangePoint: Marked as a change because one of 3/28/2023 11:38:05 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 29.02423597471476 (T) = (0 -58237781.48386553) / Math.Sqrt((2371321371901.055 / (299)) + (1950837502340.0625 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10932349178491038 = (65386008.215905115 - 58237781.48386553) / 65386008.215905115 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_DateTimes.WriteDateTimes(Formatted: True, SkipValidation: True)

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 63.083842458333336 < 69.6205037125.
IsChangePoint: Marked as a change because one of 3/29/2023 3:17:08 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 31.73193557814215 (T) = (0 -63368471.76131199) / Math.Sqrt((3029916669927.075 / (299)) + (2002811330193.612 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.11264295684202584 = (71412597.95019247 - 63368471.76131199) / 71412597.95019247 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_DateTimes.WriteDateTimes(Formatted: False, SkipValidation: True)

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 58.796892799999995 < 61.42725172002976.
IsChangePoint: Marked as a change because one of 3/15/2023 8:07:42 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 35.60735015459926 (T) = (0 -57059232.64736722) / Math.Sqrt((2083247408404.535 / (299)) + (1117046125120.7905 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.10845029835895614 = (64000058.03640595 - 57059232.64736722) / 64000058.03640595 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Text.Json.Tests.Perf_DateTimes.WriteDateTimes(Formatted: True, SkipValidation: False)

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 66.68474688461538 < 70.62883261010415.
IsChangePoint: Marked as a change because one of 3/28/2023 6:01:18 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 34.367369233845395 (T) = (0 -64959612.49710671) / Math.Sqrt((3310099918804.966 / (299)) + (1484884666157.1123 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10691565238694373 = (72736256.8504577 - 64959612.49710671) / 72736256.8504577 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
CreateNew - Duration of single invocation 44.22 μs 12.71 μs 0.29 0.00 True
CreateNew - Duration of single invocation 44.40 μs 12.77 μs 0.29 0.00 True
CreateNew - Duration of single invocation 44.75 μs 12.53 μs 0.28 0.01 True
CreateNew - Duration of single invocation 44.31 μs 12.56 μs 0.28 0.01 True

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile*' --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.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile*' --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.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile*' --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.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile*' --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

Histogram

System.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile.CreateNew(capacity: 10000)

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 12.706219588122606 < 42.75712641513748.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 503.2863665531086 (T) = (0 -12656.134293947252) / Math.Sqrt((1075776.5189956133 / (299)) + (24838.7621155476 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.722527956381694 = (45612.2863006595 - 12656.134293947252) / 45612.2863006595 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile.CreateNew(capacity: 10000000)

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 12.765560937316584 < 42.330124172401604.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 486.1694227952725 (T) = (0 -12708.062466079324) / Math.Sqrt((1116114.228474435 / (299)) + (30698.113722643164 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.7214966816363257 = (45629.842189115065 - 12708.062466079324) / 45629.842189115065 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile.CreateNew(capacity: 100000)

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 12.527844567256164 < 42.43665559279338.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 483.5864077443657 (T) = (0 -12657.711524973212) / Math.Sqrt((1095558.393942476 / (299)) + (36358.280397161376 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.7225543793543029 = (45622.31508832258 - 12657.711524973212) / 45622.31508832258 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.IO.MemoryMappedFiles.Tests.Perf_MemoryMappedFile.CreateNew(capacity: 1000000)

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 12.564214220595046 < 42.218622214114006.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 474.89953158718413 (T) = (0 -12629.621137358823) / Math.Sqrt((1170128.8077070778 / (299)) + (33297.49374295278 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.7228981509376403 = (45577.541904156045 - 12629.621137358823) / 45577.541904156045 is greater than 0.05.
IsChangeEdgeDetector: Marked as 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Collections.Tests.Perf_BitArray

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
BitArraySetAll - Duration of single invocation 554.74 ns 392.99 ns 0.71 0.11 True
BitArrayLengthValueCtor - Duration of single invocation 216.41 ns 182.87 ns 0.85 0.05 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Collections.Tests.Perf_BitArray*' --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.Collections.Tests.Perf_BitArray*' --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.Collections.Tests.Perf_BitArray*' --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.Collections.Tests.Perf_BitArray*' --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

Histogram

System.Collections.Tests.Perf_BitArray.BitArraySetAll(Size: 512)

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 392.98788097261206 < 517.7922526390832.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 63.62400365370228 (T) = (0 -401.9138016527343) / Math.Sqrt((241.98665689896515 / (299)) + (175.98067283774074 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2719109909247031 = (552.0119060211903 - 401.9138016527343) / 552.0119060211903 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Collections.Tests.Perf_BitArray.BitArrayLengthValueCtor(Size: 512)

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 182.8740203973367 < 202.08010002366066.
IsChangePoint: Marked as a change because one of 3/16/2023 10:20:48 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 48.84130826857 (T) = (0 -184.26171224666007) / Math.Sqrt((29.790168595892833 / (299)) + (9.613040024677204 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.13830515617737082 = (213.83638717070983 - 184.26171224666007) / 213.83638717070983 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Tests.Perf_Array

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Reverse - Duration of single invocation 923.99 ns 804.31 ns 0.87 0.06 True

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Tests.Perf_Array*' --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.Tests.Perf_Array*' --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.Tests.Perf_Array*' --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.Tests.Perf_Array*' --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

Histogram

System.Tests.Perf_Array.Reverse

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 804.3116555800951 < 878.6889086278361.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 14.629422261768347 (T) = (0 -785.4187328329101) / Math.Sqrt((415.20781290303603 / (299)) + (3548.443744645881 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.15520841682268036 = (929.7189371595014 - 785.4187328329101) / 929.7189371595014 is greater than 0.05.
IsChangeEdgeDetector: Marked as 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Tests.Perf_GC<Char>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
AllocateUninitializedArray - Duration of single invocation 190.24 ns 144.02 ns 0.76 0.04 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Tests.Perf_GC&lt;Char&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.Tests.Perf_GC&lt;Char&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.Tests.Perf_GC&lt;Char&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.Tests.Perf_GC&lt;Char&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

Histogram

System.Tests.Perf_GC<Char>.AllocateUninitializedArray(length: 1000, pinned: False)

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 144.01917277259875 < 146.91539007730316.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 23.98965099266915 (T) = (0 -146.72010757072965) / Math.Sqrt((29.0820812978317 / (299)) + (4.39175293575709 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.0706180468260499 = (157.86847062143073 - 146.72010757072965) / 157.86847062143073 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Globalization.Tests.Perf_DateTimeCultureInfo

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToString - Duration of single invocation 2.50 μs 1.64 μs 0.65 0.03 False
ToString - Duration of single invocation 2.22 μs 1.54 μs 0.69 0.02 True
ToString - Duration of single invocation 2.40 μs 1.66 μs 0.69 0.03 False
ToString - Duration of single invocation 2.40 μs 1.64 μs 0.69 0.03 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Globalization.Tests.Perf_DateTimeCultureInfo*' --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.Globalization.Tests.Perf_DateTimeCultureInfo*' --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.Globalization.Tests.Perf_DateTimeCultureInfo*' --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.Globalization.Tests.Perf_DateTimeCultureInfo*' --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

Histogram

System.Globalization.Tests.Perf_DateTimeCultureInfo.ToString(culturestring: fr)

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 1.6358678654525733 < 2.3383984623170977.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 39.080960540874834 (T) = (0 -1697.051611712791) / Math.Sqrt((11207.393841690253 / (299)) + (2335.250543644105 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.1889559782981349 = (2092.4284826756507 - 1697.051611712791) / 2092.4284826756507 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Globalization.Tests.Perf_DateTimeCultureInfo.ToString(culturestring: da)

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 1.5375988024557972 < 2.112371305423423.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 50.323413923686786 (T) = (0 -1581.4589495165828) / Math.Sqrt((5289.318798925455 / (299)) + (2409.942796335774 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2245620340550329 = (2039.4396701860985 - 1581.4589495165828) / 2039.4396701860985 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Globalization.Tests.Perf_DateTimeCultureInfo.ToString(culturestring: ja)

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 1.6562824790628647 < 2.3189019818566323.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 40.030233236211544 (T) = (0 -1663.074762604694) / Math.Sqrt((14157.095105334294 / (299)) + (1449.0084629232874 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.1829248985961193 = (2035.4001238652786 - 1663.074762604694) / 2035.4001238652786 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Globalization.Tests.Perf_DateTimeCultureInfo.ToString(culturestring: )

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 1.642054631205472 < 2.319839858210168.
IsChangePoint: Marked as a change because one of 3/16/2023 9:46:36 PM, 4/9/2023 4:09:45 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 35.29715032951981 (T) = (0 -1701.4214824049739) / Math.Sqrt((11667.43797411637 / (299)) + (3041.991394501123 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.18595039879926448 = (2090.071022570801 - 1701.4214824049739) / 2090.071022570801 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Single>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.50 ns 3.11 ns 0.69 0.55 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Single>.ZeroBenchmark

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 3.1111889399795403 < 4.205525980284338.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/5/2023 6:15:03 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 14.094901686156309 (T) = (0 -2.947774132925984) / Math.Sqrt((0.5085426892162123 / (299)) + (0.06876282485990896 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.22319601239673253 = (3.794746396733848 - 2.947774132925984) / 3.794746396733848 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<SByte>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.43 ns 2.96 ns 0.67 0.53 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;SByte&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;SByte&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;SByte&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;SByte&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<SByte>.ZeroBenchmark

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 2.955226462959653 < 4.207635805806058.
IsChangePoint: Marked as a change because one of 2/17/2023 6:13:48 AM, 2/17/2023 5:12:50 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.751216273071174 (T) = (0 -2.902186123119083) / Math.Sqrt((1.3390840330190583 / (299)) + (0.11726520576840842 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2446949579713157 = (3.842402687163402 - 2.902186123119083) / 3.842402687163402 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Int

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.49 ns 2.90 ns 0.65 0.48 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Int*' --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.Runtime.Intrinsics.Tests.Perf_Vector128Int*' --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.Runtime.Intrinsics.Tests.Perf_Vector128Int*' --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.Runtime.Intrinsics.Tests.Perf_Vector128Int*' --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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Int.ZeroBenchmark

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 2.9026761805977044 < 4.088774637743976.
IsChangePoint: Marked as a change because one of 2/17/2023 6:13:48 AM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 10.897846941456306 (T) = (0 -2.9234401558875307) / Math.Sqrt((0.43959686029592027 / (299)) + (0.16882862656551584 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2245338180172221 = (3.769913148775399 - 2.9234401558875307) / 3.769913148775399 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

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

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
LessThanOrEqualAnyBenchmark - Duration of single invocation 24.10 ns 21.02 ns 0.87 0.26 False
ZeroBenchmark - Duration of single invocation 4.60 ns 2.81 ns 0.61 0.54 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

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;Double&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;Double&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;Double&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;Double&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

Histogram

System.Numerics.Tests.Perf_VectorOf<Double>.LessThanOrEqualAnyBenchmark

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 21.0215757300616 < 21.43404480738389.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 12.513173231115072 (T) = (0 -21.135738827278185) / Math.Sqrt((2.805779488815914 / (299)) + (1.1404534687927188 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.10611618601644052 = (23.644838956292947 - 21.135738827278185) / 23.644838956292947 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorOf<Double>.ZeroBenchmark

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 2.8099303168236993 < 4.218986732877996.
IsChangePoint: Marked as a change because one of 2/20/2023 12:02:23 AM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 16.65367234877352 (T) = (0 -2.854396766039862) / Math.Sqrt((0.6506980672449749 / (299)) + (0.05042916595301451 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2576617503890885 = (3.845143056465113 - 2.854396766039862) / 3.845143056465113 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int16>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.51 ns 2.98 ns 0.66 0.58 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int16&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int16&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int16&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int16&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int16>.ZeroBenchmark

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 2.9800868477302975 < 4.633342527204644.
IsChangePoint: Marked as a change because one of 2/16/2023 1:00:08 AM, 3/13/2023 11:01:07 PM, 4/5/2023 11:58:06 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 12.480804337591021 (T) = (0 -2.8531963632407846) / Math.Sqrt((0.6299703405128128 / (299)) + (0.1408416030580996 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.2517104324945945 = (3.8129575596684684 - 2.8531963632407846) / 3.8129575596684684 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Double>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ZeroBenchmark - Duration of single invocation 4.50 ns 2.91 ns 0.65 0.54 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Double&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Double&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Double&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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Double&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

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Double>.ZeroBenchmark

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 2.909969640804563 < 4.07634361830989.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/6/2023 6:23:49 AM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 14.338756030638216 (T) = (0 -2.904463785752361) / Math.Sqrt((0.39766097171509557 / (299)) + (0.09687222377195549 / (36))) is greater than 1.9671134448822736 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (36) - 2, .975) and 0.2384108428105664 = (3.8136884675078435 - 2.904463785752361) / 3.8136884675078435 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Threading.Tests.Perf_Volatile

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Write_double - Duration of single invocation 6.63 ns 5.41 ns 0.82 0.32 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Threading.Tests.Perf_Volatile*' --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.Threading.Tests.Perf_Volatile*' --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.Threading.Tests.Perf_Volatile*' --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.Threading.Tests.Perf_Volatile*' --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

Histogram

System.Threading.Tests.Perf_Volatile.Write_double

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 5.414298349969732 < 6.300405038274694.
IsChangePoint: Marked as a change because one of 2/15/2023 6:58:43 PM, 3/14/2023 5:06:59 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 14.323436969659607 (T) = (0 -5.286734089945418) / Math.Sqrt((0.21036920917137578 / (299)) + (0.2733002395362305 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.19594044933390745 = (6.575052911896668 - 5.286734089945418) / 6.575052911896668 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 x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Memory.MemoryMarshal<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
CastToByte - Duration of single invocation 27.04 ns 23.63 ns 0.87 0.21 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Memory.MemoryMarshal&lt;Int32&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.Memory.MemoryMarshal&lt;Int32&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.Memory.MemoryMarshal&lt;Int32&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.Memory.MemoryMarshal&lt;Int32&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

Histogram

System.Memory.MemoryMarshal<Int32>.CastToByte

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 23.632775030130798 < 25.728107332574986.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 32.546179314937206 (T) = (0 -23.896348583722578) / Math.Sqrt((0.9920715018774063 / (299)) + (0.21133759958990422 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.11459117797283982 = (26.989056342370112 - 23.896348583722578) / 26.989056342370112 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 Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline e7ed417c2bf4a24a0694a387f33ba843a2318fdc
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Improvements in System.Tests.Perf_String

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToUpperInvariant - Duration of single invocation 361.16 ns 334.32 ns 0.93 0.17 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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.Tests.Perf_String*' --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.Tests.Perf_String*' --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.Tests.Perf_String*' --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.Tests.Perf_String*' --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

Histogram

System.Tests.Perf_String.ToUpperInvariant(s: "TeSt")

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 334.3192140146132 < 340.6135402641473.
IsChangePoint: Marked as a change because one of 2/14/2023 9:29:25 PM, 3/16/2023 9:46:36 PM, 4/12/2023 4:44:56 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 4:44:56 PM and 4/25/2023 3:46:39 AM.
IsImprovementStdDev: Marked as improvement because 19.19927983280111 (T) = (0 -337.7885851234336) / Math.Sqrt((176.67121316386715 / (299)) + (62.44469558811636 / (37))) is greater than 1.9670919629190173 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (37) - 2, .975) and 0.07901734389615166 = (366.76975715528044 - 337.7885851234336) / 366.76975715528044 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

Some of the improvements could be dotnet/runtime#84963.

@kotlarmilos kotlarmilos mentioned this issue May 9, 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
arch-x64 branch-refs/heads/main kind-micro_mono mono-interpreter os-linux perf-improvement runtime-mono
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@kotlarmilos