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.

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

[Perf] Linux/arm64: 16 Improvements on 4/26/2023 2:24:59 PM #17314

Closed
performanceautofiler bot opened this issue May 2, 2023 · 1 comment
Closed

[Perf] Linux/arm64: 16 Improvements on 4/26/2023 2:24:59 PM #17314

performanceautofiler bot opened this issue May 2, 2023 · 1 comment

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented May 2, 2023

Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline 876ab8bc7ae8f77a9dd8e630a895a77d344ba005
Compare 678b4c842c647f3b6a841b24a049c5f00aebf65a
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:true, MonoInterpreter:false, RunKind:micro_mono

Improvements in System.Numerics.Tests.Perf_VectorConvert

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Convert_int_float - Duration of single invocation 17.45 μs 2.48 μs 0.14 0.02 False
Convert_ulong_double - Duration of single invocation 49.11 μs 2.50 μs 0.05 0.03 True
Convert_long_double - Duration of single invocation 49.60 μs 2.56 μs 0.05 0.01 True
Convert_uint_float - Duration of single invocation 51.64 μs 2.48 μs 0.05 0.02 True
Convert_double_long - Duration of single invocation 10.33 μs 2.54 μs 0.25 0.02 False
Convert_double_ulong - Duration of single invocation 46.06 μs 2.48 μs 0.05 0.48 False
Convert_float_uint - Duration of single invocation 64.62 μs 2.51 μs 0.04 0.53 False
Convert_float_int - Duration of single invocation 16.24 μs 2.49 μs 0.15 0.01 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 (Files either built locally (with build.(sh/cmd) or downloaded from payload above (if same system setup) (in this order))

  • 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 $RunArch -framework net8.0
  • AOT MONO build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release /p:CrossBuild=false /p:MonoLLVMUseCxx11Abi=false
  • 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 aot directory 
mkdir -p $RunDir/artifacts/bin/aot/sgen
mkdir -p $RunDir/artifacts/bin/aot/pack
cp -r $RunDir/artifacts/obj/mono/$RunOS.$RunArch.Release/mono/* $RunDir/artifacts/bin/aot/sgen
cp -r $RunDir/artifacts/bin/microsoft.netcore.app.runtime.$RunOS-$RunArch/Release/* $RunDir/artifacts/bin/aot/pack

# 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

# 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_VectorConvert*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir/artifacts/bin/aot/sgen/mini/mono-sgen --customruntimepack $RunDir/artifacts/bin/aot/pack --aotcompilermode llvm --logBuildOutput --generateBinLog"

# 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_VectorConvert* --anyCategories Libraries Runtime " --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir/artifacts/bin/aot/sgen/mini/mono-sgen --customruntimepack $RunDir/artifacts/bin/aot/pack --aotcompilermode llvm --logBuildOutput --generateBinLog " --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages --buildTimeout 1200

Windows

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

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

# Create aot directory
mkdir $RunDir\artifacts\bin\aot\sgen
mkdir $RunDir\artifacts\bin\aot\pack
xcopy $RunDir\artifacts\obj\mono\$RunOS.$RunArch.Release\mono $RunDir\artifacts\bin\aot\sgen\ /e /y
xcopy $RunDir\artifacts\bin\microsoft.netcore.app.runtime.$RunOSId-$RunArch\Release $RunDir\artifacts\bin\aot\pack\ /e /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

# 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_VectorConvert*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir\artifacts\bin\aot\sgen\mini\mono-sgen.exe --customruntimepack $RunDir\artifacts\bin\aot\pack --aotcompilermode llvm --logBuildOutput --generateBinLog"

# 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_VectorConvert* --anyCategories Libraries Runtime " --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir\artifacts\bin\aot\sgen\mini\mono-sgen.exe --customruntimepack $RunDir\artifacts\bin\aot\pack -aotcompilermode llvm --logBuildOutput --generateBinLog " --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages --buildTimeout 1200

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_VectorConvert.Convert_int_float


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 2.4846373422734107 < 16.56548743892616.
IsChangePoint: Marked as a change because one of 3/15/2023 10:57:55 AM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 78.10810147330687 (T) = (0 -2514.598546262692) / Math.Sqrt((23190067.823539596 / (299)) + (475.63461835227565 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.8963937603818922 = (24270.724963394994 - 2514.598546262692) / 24270.724963394994 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Convert_ulong_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 2.4972415211384726 < 46.88561119540843.
IsChangePoint: Marked as a change because one of 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 262.8656269234597 (T) = (0 -2518.12558962505) / Math.Sqrt((11535782.793396229 / (299)) + (313.85745268934477 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9535071013114972 = (54161.5098360765 - 2518.12558962505) / 54161.5098360765 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Convert_long_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 2.5588029657873044 < 47.71936258188986.
IsChangePoint: Marked as a change because one of 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 303.3493906185897 (T) = (0 -2520.086886261114) / Math.Sqrt((8674516.275041133 / (299)) + (581.1931882899296 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9535179382250656 = (54216.33184998003 - 2520.086886261114) / 54216.33184998003 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Convert_uint_float


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 2.4826424358974357 < 48.81169660756142.
IsChangePoint: Marked as a change because one of 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 302.23890742367007 (T) = (0 -2511.2991937237457) / Math.Sqrt((8469404.19979268 / (299)) + (405.0796756618712 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9529701937932723 = (53398.03406131194 - 2511.2991937237457) / 53398.03406131194 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Convert_double_long


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 2.5410375084471304 < 9.796246285280622.
IsChangePoint: Marked as a change because one of 3/15/2023 10:57:55 AM, 3/24/2023 12:04:20 PM, 4/3/2023 11:26:06 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 51.80203151199185 (T) = (0 -2528.8607069503328) / Math.Sqrt((23929930.438643172 / (299)) + (810.9160543121925 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.8528674494059064 = (17187.635888450703 - 2528.8607069503328) / 17187.635888450703 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Convert_double_ulong


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 2.4817650439781698 < 30.82583755962034.
IsChangePoint: Marked as a change because one of 4/3/2023 11:26:06 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 95.89137730804153 (T) = (0 -2522.5586815525035) / Math.Sqrt((52055460.662766464 / (299)) + (850.4622530234316 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9406993838142729 = (42538.490218245766 - 2522.5586815525035) / 42538.490218245766 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Convert_float_uint


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 2.5127709099338813 < 64.27937211323243.
IsChangePoint: Marked as a change because one of 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 114.7739635506073 (T) = (0 -2512.4191486670393) / Math.Sqrt((128968966.98370066 / (299)) + (418.0111704839969 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9677453943646936 = (77893.34574647271 - 2512.4191486670393) / 77893.34574647271 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Convert_float_int


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 2.4853409678840297 < 15.478064479983608.
IsChangePoint: Marked as a change because one of 3/15/2023 10:57:55 AM, 3/24/2023 12:04:20 PM, 4/3/2023 11:26:06 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 77.7407338895983 (T) = (0 -2509.980679646885) / Math.Sqrt((20347847.58743828 / (299)) + (416.9944291865391 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.8898815064973525 = (22793.452759926644 - 2509.980679646885) / 22793.452759926644 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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


Run Information

Name Value
Architecture arm64
OS ubuntu 20.04
Queue AmpereUbuntu
Baseline 876ab8bc7ae8f77a9dd8e630a895a77d344ba005
Compare 678b4c842c647f3b6a841b24a049c5f00aebf65a
Diff Diff
Configs CompilationMode:tiered, LLVM:false, MonoAOT:true, MonoInterpreter:false, RunKind:micro_mono

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ConvertFloatToUIntBenchmark - Duration of single invocation 62.24 ns 1.31 ns 0.02 0.56 False
ConvertIntToFloatBenchmark - Duration of single invocation 20.17 ns 1.10 ns 0.05 0.09 True
ConvertFloatToIntBenchmark - Duration of single invocation 19.96 ns 1.44 ns 0.07 0.07 True
ConvertULongToDoubleBenchmark - Duration of single invocation 16.46 ns 1.23 ns 0.07 0.07 True
ConvertLongToDoubleBenchmark - Duration of single invocation 16.52 ns 1.18 ns 0.07 0.08 True
ConvertDoubleToLongBenchmark - Duration of single invocation 14.38 ns 1.41 ns 0.10 0.08 True
ConvertUIntToFloatBenchmark - Duration of single invocation 18.26 ns 1.54 ns 0.08 0.10 True
ConvertDoubleToULongBenchmark - Duration of single invocation 46.29 ns 1.30 ns 0.03 0.59 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 (Files either built locally (with build.(sh/cmd) or downloaded from payload above (if same system setup) (in this order))

  • 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 $RunArch -framework net8.0
  • AOT MONO build extracted to runtime/artifacts/bin/mono/$RunOS.$RunArch.Release, build instructions: MONO README args: -arch $RunArch -os $RunOS -s mono+libs+host+packs -c Release /p:CrossBuild=false /p:MonoLLVMUseCxx11Abi=false
  • 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 aot directory 
mkdir -p $RunDir/artifacts/bin/aot/sgen
mkdir -p $RunDir/artifacts/bin/aot/pack
cp -r $RunDir/artifacts/obj/mono/$RunOS.$RunArch.Release/mono/* $RunDir/artifacts/bin/aot/sgen
cp -r $RunDir/artifacts/bin/microsoft.netcore.app.runtime.$RunOS-$RunArch/Release/* $RunDir/artifacts/bin/aot/pack

# 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

# 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_Vector128*' --bdn-artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir/artifacts/bin/aot/sgen/mini/mono-sgen --customruntimepack $RunDir/artifacts/bin/aot/pack --aotcompilermode llvm --logBuildOutput --generateBinLog"

# 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_Vector128* --anyCategories Libraries Runtime " --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir/artifacts/bin/aot/sgen/mini/mono-sgen --customruntimepack $RunDir/artifacts/bin/aot/pack --aotcompilermode llvm --logBuildOutput --generateBinLog " --artifacts $RunDir/artifacts/BenchmarkDotNet.Artifacts --packages $RunDir/performance/artifacts/packages --buildTimeout 1200

Windows

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

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

# Create aot directory
mkdir $RunDir\artifacts\bin\aot\sgen
mkdir $RunDir\artifacts\bin\aot\pack
xcopy $RunDir\artifacts\obj\mono\$RunOS.$RunArch.Release\mono $RunDir\artifacts\bin\aot\sgen\ /e /y
xcopy $RunDir\artifacts\bin\microsoft.netcore.app.runtime.$RunOSId-$RunArch\Release $RunDir\artifacts\bin\aot\pack\ /e /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

# 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_Vector128*' --bdn-artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --bdn-arguments="--anyCategories Libraries Runtime  --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir\artifacts\bin\aot\sgen\mini\mono-sgen.exe --customruntimepack $RunDir\artifacts\bin\aot\pack --aotcompilermode llvm --logBuildOutput --generateBinLog"

# 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_Vector128* --anyCategories Libraries Runtime " --category-exclusion-filter NoAOT NoWASM --runtimes monoaotllvm --aotcompilerpath $RunDir\artifacts\bin\aot\sgen\mini\mono-sgen.exe --customruntimepack $RunDir\artifacts\bin\aot\pack -aotcompilermode llvm --logBuildOutput --generateBinLog " --artifacts $RunDir\artifacts\BenchmarkDotNet.Artifacts --packages $RunDir\performance\artifacts\packages --buildTimeout 1200

Payloads

Baseline
Compare

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertFloatToUIntBenchmark


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 1.307919047569265 < 58.541566525919436.
IsChangePoint: Marked as a change because one of 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 93.0347697718243 (T) = (0 -1.2879504639541295) / Math.Sqrt((158.92850720000544 / (299)) + (0.03549659622327608 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9813974871947772 = (69.2352951152129 - 1.2879504639541295) / 69.2352951152129 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertIntToFloatBenchmark


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 1.1029102825918875 < 19.21183886530774.
IsChangePoint: Marked as a change because one of 1/5/2023 4:05:11 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 189.17269534499496 (T) = (0 -1.3538401781378184) / Math.Sqrt((0.25083222198416033 / (299)) + (0.1609273990059119 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9309462190808565 = (19.605590890425795 - 1.3538401781378184) / 19.605590890425795 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertFloatToIntBenchmark


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 1.4374260592550034 < 16.70914088320526.
IsChangePoint: Marked as a change because one of 1/5/2023 4:05:11 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 442.3302219732615 (T) = (0 -1.2605507964015794) / Math.Sqrt((0.06635009954776301 / (299)) + (0.021794926661106332 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9284867613996366 = (17.62681737077941 - 1.2605507964015794) / 17.62681737077941 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertULongToDoubleBenchmark


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 1.2308677497740337 < 15.635453285461146.
IsChangePoint: Marked as a change because one of 1/5/2023 4:05:11 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 322.4419054190712 (T) = (0 -1.2339804356195576) / Math.Sqrt((0.31536991223150546 / (299)) + (0.018400044391944082 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9215888911295653 = (15.737316477166102 - 1.2339804356195576) / 15.737316477166102 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertLongToDoubleBenchmark


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 1.175714163738996 < 15.620014750201904.
IsChangePoint: Marked as a change because one of 1/5/2023 4:05:11 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 367.1554844829165 (T) = (0 -1.2423904187947978) / Math.Sqrt((0.34079853245674363 / (299)) + (0.007912642478088797 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9209998275677814 = (15.726426671546289 - 1.2423904187947978) / 15.726426671546289 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertDoubleToLongBenchmark


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 1.4146726566663905 < 13.611756835984888.
IsChangePoint: Marked as a change because one of 1/5/2023 4:05:11 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 104.63421909971076 (T) = (0 -1.3571262116623495) / Math.Sqrt((0.6839280163711153 / (299)) + (0.2080048290526797 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.8986811940418433 = (13.3946131601948 - 1.3571262116623495) / 13.3946131601948 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertUIntToFloatBenchmark


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 1.544609886822255 < 17.34664308663408.
IsChangePoint: Marked as a change because one of 1/5/2023 4:05:11 PM, 3/6/2023 4:10:52 PM, 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 190.621083821673 (T) = (0 -1.272282430181418) / Math.Sqrt((1.4960468345455693 / (299)) + (0.03446400351941649 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9252099811793558 = (17.011393368311758 - 1.272282430181418) / 17.011393368311758 is greater than 0.05.
IsChangeEdgeDetector: Marked as regression because Edge Detector said so.

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128.ConvertDoubleToULongBenchmark


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 1.2981744789676817 < 39.74411363015776.
IsChangePoint: Marked as a change because one of 4/26/2023 5:34:01 AM, 5/2/2023 12:51:17 PM falls between 4/23/2023 8:38:14 PM and 5/2/2023 12:51:17 PM.
IsImprovementStdDev: Marked as improvement because 101.63779767527312 (T) = (0 -1.24291572484113) / Math.Sqrt((40.526871386161574 / (299)) + (0.009814169344146874 / (19))) is greater than 1.9674995188240387 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (19) - 2, .975) and 0.9679107297147606 = (38.733062914579776 - 1.24291572484113) / 38.733062914579776 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

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant