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: 14 Regressions on 7/12/2023 7:15:00 AM #19919

Closed
performanceautofiler bot opened this issue Jul 18, 2023 · 6 comments
Closed

[Perf] Linux/x64: 14 Regressions on 7/12/2023 7:15:00 AM #19919

performanceautofiler bot opened this issue Jul 18, 2023 · 6 comments
Labels
arch-x64 branch-refs/heads/main kind-micro_mono LinkedFromMaster Is this issue linked from the master perf issue mono-interpreter os-linux perf-regression PGO Applied if there were any profile guided optimization updates in the observed interval. runtime-mono

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Jul 18, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 83bf4b66ed3de5905b2e410a194a0f5f3d924c3d
Compare d9f6b98de38081b71fa9b5a463f302288546435d
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Regressions in System.Tests.Perf_Double

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
TryParse - Duration of single invocation
📝 - Benchmark Source		 1.32 μs 1.43 μs 1.08 0.08 False
Parse - Duration of single invocation
📝 - Benchmark Source		 654.25 ns 778.31 ns 1.19 0.07 False
TryParse - Duration of single invocation
📝 - Benchmark Source		 1.31 μs 1.49 μs 1.14 0.07 False
Parse - Duration of single invocation
📝 - Benchmark Source		 1.32 μs 1.44 μs 1.09 0.05 False
TryParse - Duration of single invocation
📝 - Benchmark Source		 678.50 ns 770.23 ns 1.14 0.07 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

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_Double*' --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_Double*' --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_Double*' --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_Double*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Tests.Perf_Double.TryParse(value: "-1.7976931348623157e+308")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.4283134987093364 > 1.404487021123169.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 5/26/2023 5:44:43 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -14.067276825180706 (T) = (0 -1428.106638464281) / Math.Sqrt((441.9537241284203 / (21)) + (179.5826920356992 / (7))) is less than -2.055529438636913 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (7) - 2, .025) and -0.07217270481643573 = (1331.974440357334 - 1428.106638464281) / 1331.974440357334 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Double.Parse(value: "12345")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 778.3109760841621 > 688.6149457579953.
IsChangePoint: Marked as a change because one of 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -12.23658529417109 (T) = (0 -767.9255428236728) / Math.Sqrt((458.8351216028941 / (21)) + (310.7041143005677 / (7))) is less than -2.055529438636913 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (7) - 2, .025) and -0.14900812279605108 = (668.3377842055339 - 767.9255428236728) / 668.3377842055339 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Double.TryParse(value: "1.7976931348623157e+308")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.492319784238435 > 1.3781275268458926.
IsChangePoint: Marked as a change because one of 5/10/2023 2:00:36 PM, 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -5.682934905281725 (T) = (0 -1428.8176468827073) / Math.Sqrt((518.2335862651872 / (21)) + (2394.015732077515 / (7))) is less than -2.055529438636913 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (7) - 2, .025) and -0.08244110113960654 = (1319.9957442288837 - 1428.8176468827073) / 1319.9957442288837 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Double.Parse(value: "-1.7976931348623157e+308")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.4405998906580748 > 1.3800986139191427.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 5/10/2023 6:18:38 PM, 7/11/2023 8:33:01 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -6.549582382735843 (T) = (0 -1424.8359957005018) / Math.Sqrt((1207.8847712398183 / (18)) + (1379.8277354178404 / (11))) is less than -2.051830516474954 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (18) + (11) - 2, .025) and -0.06812983122294863 = (1333.9539389786958 - 1424.8359957005018) / 1333.9539389786958 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Double.TryParse(value: "12345")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 770.2342306770473 > 694.2220032779679.
IsChangePoint: Marked as a change because one of 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -21.98709931232667 (T) = (0 -772.2271343345261) / Math.Sqrt((168.6746159646895 / (21)) + (105.08105353800386 / (7))) is less than -2.055529438636913 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (7) - 2, .025) and -0.15831653537631882 = (666.680575429791 - 772.2271343345261) / 666.680575429791 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
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 83bf4b66ed3de5905b2e410a194a0f5f3d924c3d
Compare d9f6b98de38081b71fa9b5a463f302288546435d
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Regressions in System.Tests.Perf_Single

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
Parse - Duration of single invocation
📝 - Benchmark Source		 652.83 ns 776.09 ns 1.19 0.08 False
TryParse - Duration of single invocation
📝 - Benchmark Source		 673.24 ns 772.03 ns 1.15 0.09 False
Parse - Duration of single invocation
📝 - Benchmark Source		 983.38 ns 1.12 μs 1.14 0.07 False
TryParse - Duration of single invocation
📝 - Benchmark Source		 1.01 μs 1.11 μs 1.09 0.07 False
Parse - Duration of single invocation
📝 - Benchmark Source		 998.47 ns 1.15 μs 1.15 0.06 False
TryParse - Duration of single invocation
📝 - Benchmark Source		 996.82 ns 1.12 μs 1.12 0.07 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

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_Single*' --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_Single*' --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_Single*' --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_Single*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Tests.Perf_Single.Parse(value: "12345")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 776.0932317703125 > 704.9322950498149.
IsChangePoint: Marked as a change because one of 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -22.364599353950155 (T) = (0 -782.9983930504808) / Math.Sqrt((259.7258235265031 / (21)) + (99.82029583698599 / (8))) is less than -2.051830516474954 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (8) - 2, .025) and -0.16600624055240706 = (671.5216143950718 - 782.9983930504808) / 671.5216143950718 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Single.TryParse(value: "12345")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 772.0319310040405 > 699.3400812757429.
IsChangePoint: Marked as a change because one of 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -21.48071236296315 (T) = (0 -787.0136719641645) / Math.Sqrt((236.51532775250817 / (21)) + (140.92185317931308 / (8))) is less than -2.051830516474954 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (8) - 2, .025) and -0.1718831310259696 = (671.5803403323533 - 787.0136719641645) / 671.5803403323533 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Single.Parse(value: "3.4028235E+38")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.1229082434515634 > 1027.808843632553.
IsChangePoint: Marked as a change because one of 5/10/2023 12:32:34 AM, 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -15.366255475709803 (T) = (0 -1120.119588638584) / Math.Sqrt((301.06533340480496 / (21)) + (419.28001657656336 / (7))) is less than -2.055529438636913 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (7) - 2, .025) and -0.13403924479156226 = (987.7255957261543 - 1120.119588638584) / 987.7255957261543 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Single.TryParse(value: "3.4028235E+38")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.1068694844980786 > 1036.0246867091987.
IsChangePoint: Marked as a change because one of 5/10/2023 2:00:36 PM, 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -18.833953240172935 (T) = (0 -1109.263060964777) / Math.Sqrt((210.56575582699378 / (21)) + (219.25774856852763 / (7))) is less than -2.055529438636913 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (7) - 2, .025) and -0.12256087248773327 = (988.154039706117 - 1109.263060964777) / 988.154039706117 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Single.Parse(value: "-3.4028235E+38")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.1463628462353526 > 1049.083427715319.
IsChangePoint: Marked as a change because one of 5/10/2023 2:00:36 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -21.53241259488577 (T) = (0 -1144.088590498105) / Math.Sqrt((221.58077121821447 / (21)) + (296.32219060310575 / (8))) is less than -2.051830516474954 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (8) - 2, .025) and -0.1492101153231692 = (995.5434391354762 - 1144.088590498105) / 995.5434391354762 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Single.TryParse(value: "-3.4028235E+38")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.1166115019202196 > 1040.215621396703.
IsChangePoint: Marked as a change because one of 4/12/2023 4:44:56 PM, 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -20.081663758962534 (T) = (0 -1139.0482983965712) / Math.Sqrt((335.3556259693156 / (21)) + (282.2598588647695 / (9))) is less than -2.048407141790485 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (9) - 2, .025) and -0.138034705592966 = (1000.8906519270668 - 1139.0482983965712) / 1000.8906519270668 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
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 83bf4b66ed3de5905b2e410a194a0f5f3d924c3d
Compare d9f6b98de38081b71fa9b5a463f302288546435d
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Regressions in System.Tests.Perf_Decimal

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
Parse - Duration of single invocation
📝 - Benchmark Source		 872.37 ns 1.03 μs 1.18 0.04 False
TryParse - Duration of single invocation
📝 - Benchmark Source		 869.85 ns 997.79 ns 1.15 0.07 False

graph
graph
Test Report

Repro

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

Repro Steps

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

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

Linux

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

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

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

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

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

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

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

Payloads

Baseline
Compare

System.Tests.Perf_Decimal.Parse(value: "123456.789")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 1.0320032823791305 > 904.3949509603732.
IsChangePoint: Marked as a change because one of 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -26.943591650363345 (T) = (0 -1024.3770950240116) / Math.Sqrt((511.7817955246586 / (21)) + (80.58593559685896 / (8))) is less than -2.051830516474954 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (8) - 2, .025) and -0.1825444608552322 = (866.248271361542 - 1024.3770950240116) / 866.248271361542 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Tests.Perf_Decimal.TryParse(value: "123456.789")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 997.7892325416518 > 900.2954483174825.
IsChangePoint: Marked as a change because one of 6/15/2023 3:33:58 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -24.295394202879734 (T) = (0 -1005.9802128601907) / Math.Sqrt((492.91104018827946 / (21)) + (59.01692964758417 / (7))) is less than -2.055529438636913 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (7) - 2, .025) and -0.15795845288288596 = (868.7532876120677 - 1005.9802128601907) / 868.7532876120677 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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

performanceautofiler bot commented Jul 18, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 83bf4b66ed3de5905b2e410a194a0f5f3d924c3d
Compare d9f6b98de38081b71fa9b5a463f302288546435d
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:false, MonoInterpreter:true, RunKind:micro_mono

Regressions in System.Tests.Perf_UInt32

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
ParseSpan - Duration of single invocation
📝 - Benchmark Source		 205.75 ns 223.80 ns 1.09 0.21 False

graph
Test Report

Repro

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

Repro Steps

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

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

Linux

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

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

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

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

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

# Use Interpreter
export MONO_ENV_OPTIONS="--interpreter"

# One line run:
python3 $RunDir/performance/scripts/benchmarks_ci.py --csproj $RunDir/performance/src/benchmarks/micro/MicroBenchmarks.csproj --incremental no --architecture $RunArch -f net8.0 --filter 'System.Tests.Perf_UInt32*' --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_UInt32*' --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_UInt32*' --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_UInt32*' --anyCategories Libraries Runtime " --category-exclusion-filter NoInterpreter NoMono --logBuildOutput --generateBinLog --corerun $RunDir\artifacts\dotnet-mono\shared\Microsoft.NETCore.App\8.0.0\corerun.exe --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages

Payloads

Baseline
Compare

System.Tests.Perf_UInt32.ParseSpan(value: "4294967295")

ETL Files

Histogram

Description of detection logic

IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
IsRegressionBase: Marked as regression because the compare was 5% greater than the baseline, and the value was not too small.
IsRegressionChecked: Marked as regression because the three check build points were 0.05 greater than the baseline.
IsRegressionWindowed: Marked as regression because 223.79621017638308 > 216.52668419772613.
IsChangePoint: Marked as a change because one of 4/17/2023 1:54:35 PM, 7/12/2023 2:12:47 AM, 7/17/2023 10:23:26 PM falls between 7/8/2023 10:33:35 PM and 7/17/2023 10:23:26 PM.
IsRegressionStdDev: Marked as regression because -6.882453434369375 (T) = (0 -224.2445668055403) / Math.Sqrt((37.55651907340628 / (21)) + (36.02160996318624 / (8))) is less than -2.051830516474954 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (21) + (8) - 2, .025) and -0.08340159492203945 = (206.9819426670465 - 224.2445668055403) / 206.9819426670465 is less than -0.05.
IsImprovementBase: Marked as not an improvement because the compare was not 5% less than the baseline, or the value was too small.
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

@kg please take a look at the regressions. I am not sure if the Parse methods rely on dotnet/runtime#88064.

@kg
Copy link

kg commented Jul 19, 2023

@kg please take a look at the regressions. I am not sure if the Parse methods rely on dotnet/runtime#88064.

Seems far more likely to be dotnet/runtime@1dd0fa2, iirc the Vector128<T>.Equals method doesn't get used much or at all in the BCL.

@kotlarmilos kotlarmilos mentioned this issue Jul 19, 2023
Closed
@kotlarmilos
Copy link
Member

/cc: @tannergooding @BrzVlad

@tannergooding
Copy link
Member

Same with dotnet/runtime#89103, some small regressions were expected from the change.

We added significant functionality and to avoid duplicating several thousand lines of code, the implementation was switched over to use static virtuals in interfaces so the one implementation could be shared between char (UTF-16) and byte (UTF-8).

If there's any small fix we can do to win back the perf, we should. The numbers here are a bit higher than the 4ns lost for RyuJIT, but are all still within a general 10-15% margin and don't look particularly worrisome given the scenarios shown primarily being the edge case tests covering the Min/Max (and therefore typically longest) strings.

@kotlarmilos kotlarmilos added the LinkedFromMaster Is this issue linked from the master perf issue label Dec 19, 2023
@kotlarmilos
Copy link
Member

Not planned for .NET 9.

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 LinkedFromMaster Is this issue linked from the master perf issue mono-interpreter os-linux perf-regression PGO Applied if there were any profile guided optimization updates in the observed interval. runtime-mono
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@kg @tannergooding @kotlarmilos