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/x64: 10 Regressions on 4/17/2023 6:40:18 PM #16690

Open
performanceautofiler bot opened this issue Apr 19, 2023 · 1 comment
Open

[Perf] Linux/x64: 10 Regressions on 4/17/2023 6:40:18 PM #16690

performanceautofiler bot opened this issue Apr 19, 2023 · 1 comment

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Apr 19, 2023

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Compare 9aefa9daa141bb7d9ba3f2b373d4b050c9b243fe
Diff Diff
Configs AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in Benchstone.BenchI.BubbleSort

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Test - Duration of single invocation 13.98 μs 16.66 μs 1.19 0.00 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'Benchstone.BenchI.BubbleSort*'

Payloads

Baseline
Compare

Histogram

Benchstone.BenchI.BubbleSort.Test


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 16.658781892561645 > 14.701866402841699.
IsChangePoint: Marked as a change because one of 4/17/2023 1:54:35 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -153.16241279580066 (T) = (0 -16689.517611894484) / Math.Sqrt((618.071970332376 / (34)) + (863.4151245252449 / (3))) is less than -2.0301079282477414 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (34) + (3) - 2, .025) and -0.19122848291582448 = (14010.341299968573 - 16689.517611894484) / 14010.341299968573 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.

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 c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Compare 9aefa9daa141bb7d9ba3f2b373d4b050c9b243fe
Diff Diff
Configs AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Tests.Perf_Int16

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TryParse - Duration of single invocation 32.74 ns 41.61 ns 1.27 0.63 False
Parse - Duration of single invocation 34.85 ns 55.31 ns 1.59 0.64 False
Parse - Duration of single invocation 29.62 ns 45.14 ns 1.52 0.64 False
Parse - Duration of single invocation 32.51 ns 62.35 ns 1.92 0.60 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

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_Int16*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Int16.TryParse(value: "32767")


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 41.60801255486559 > 35.383080307333366.
IsChangePoint: Marked as a change because one of 4/17/2023 1:54:35 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -4.867418475011111 (T) = (0 -42.003359650420606) / Math.Sqrt((4.703727716160876 / (33)) + (6.219593841269267 / (3))) is less than -2.0322445093148245 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (3) - 2, .025) and -0.20845021318753362 = (34.75803900901154 - 42.003359650420606) / 34.75803900901154 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.

System.Tests.Perf_Int16.Parse(value: "-32768")


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 55.310154545609116 > 36.84920275796996.
IsChangePoint: Marked as a change because one of 4/17/2023 6:40:18 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -4.94434288066354 (T) = (0 -60.040635983106384) / Math.Sqrt((9.224378537163398 / (35)) + (44.7549092610125 / (2))) is less than -2.0301079282477414 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (35) + (2) - 2, .025) and -0.6443100118549693 = (36.514182575202895 - 60.040635983106384) / 36.514182575202895 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.

System.Tests.Perf_Int16.Parse(value: "0")


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 45.137692037892045 > 31.175075851605392.
IsChangePoint: Marked as a change because one of 4/17/2023 6:40:18 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -16.136560146275635 (T) = (0 -45.962324640048685) / Math.Sqrt((5.087574564505079 / (34)) + (1.3600378570792637 / (2))) is less than -2.0322445093148245 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (34) + (2) - 2, .025) and -0.47012223074977505 = (31.26428787938776 - 45.962324640048685) / 31.26428787938776 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.

System.Tests.Perf_Int16.Parse(value: "32767")


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 62.349159704857215 > 35.05085057111022.
IsChangePoint: Marked as a change because one of 4/17/2023 6:40:18 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -14.244584051658638 (T) = (0 -60.5671271903529) / Math.Sqrt((3.874091858800674 / (34)) + (6.351279765501148 / (2))) is less than -2.0322445093148245 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (34) + (2) - 2, .025) and -0.7438709986317713 = (34.73142637148816 - 60.5671271903529) / 34.73142637148816 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.

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 c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Compare 9aefa9daa141bb7d9ba3f2b373d4b050c9b243fe
Diff Diff
Configs AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.MathBenchmarks.Double

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Max - Duration of single invocation 4.37 μs 4.72 μs 1.08 0.00 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.MathBenchmarks.Double*'

Payloads

Baseline
Compare

Histogram

System.MathBenchmarks.Double.Max


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 4.7246419905144075 > 4.590642361590055.
IsChangePoint: Marked as a change because one of 4/17/2023 1:54:35 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -60.99946862483759 (T) = (0 -4727.158473363669) / Math.Sqrt((843.8994465338607 / (34)) + (18.170442292055185 / (3))) is less than -2.0301079282477414 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (34) + (3) - 2, .025) and -0.07724313445729859 = (4388.200139929554 - 4727.158473363669) / 4388.200139929554 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.

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 c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Compare 9aefa9daa141bb7d9ba3f2b373d4b050c9b243fe
Diff Diff
Configs AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Collections.IndexerSetReverse<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
IList - Duration of single invocation 5.81 μs 6.58 μs 1.13 0.38 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.IndexerSetReverse&lt;Int32&gt;*'

Payloads

Baseline
Compare

Histogram

System.Collections.IndexerSetReverse<Int32>.IList(Size: 512)


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 6.575333374423472 > 6.14265784724513.
IsChangePoint: Marked as a change because one of 4/17/2023 1:54:35 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -6.16439574384889 (T) = (0 -6792.449780814076) / Math.Sqrt((99031.42270704478 / (33)) + (36915.49467640371 / (3))) is less than -2.0322445093148245 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (3) - 2, .025) and -0.12647941604711804 = (6029.803726595536 - 6792.449780814076) / 6029.803726595536 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.

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 c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Compare 9aefa9daa141bb7d9ba3f2b373d4b050c9b243fe
Diff Diff
Configs AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Tests.Perf_Version

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Parse3 - Duration of single invocation 161.51 ns 176.86 ns 1.10 0.27 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_Version*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Version.Parse3


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 176.86332079447192 > 171.45890788116245.
IsChangePoint: Marked as a change because one of 4/17/2023 1:54:35 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -2.0718539292452953 (T) = (0 -187.8179065847211) / Math.Sqrt((46.00785749593048 / (34)) + (273.5937390357462 / (3))) is less than -2.0301079282477414 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (34) + (3) - 2, .025) and -0.11872318359622708 = (167.88595189469936 - 187.8179065847211) / 167.88595189469936 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.

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 c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Compare 9aefa9daa141bb7d9ba3f2b373d4b050c9b243fe
Diff Diff
Configs AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Tests.Perf_Byte

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Parse - Duration of single invocation 31.81 ns 48.44 ns 1.52 0.69 False
Parse - Duration of single invocation 38.14 ns 49.27 ns 1.29 0.67 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_Byte*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Byte.Parse(value: "0")


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 48.442604495652844 > 32.13360356781855.
IsChangePoint: Marked as a change because one of 2/16/2023 3:28:45 AM, 4/17/2023 1:54:35 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -19.97950445458046 (T) = (0 -49.11717427390863) / Math.Sqrt((9.057586314823975 / (34)) + (1.5246280098980298 / (3))) is less than -2.0301079282477414 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (34) + (3) - 2, .025) and -0.5576515247569722 = (31.532838695466232 - 49.11717427390863) / 31.532838695466232 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.

System.Tests.Perf_Byte.Parse(value: "255")


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 49.27370689275745 > 34.63949147109729.
IsChangePoint: Marked as a change because one of 2/14/2023 9:29:25 PM, 4/17/2023 6:40:18 PM, 4/18/2023 4:36:42 PM falls between 4/9/2023 11:04:55 PM and 4/18/2023 4:36:42 PM.
IsRegressionStdDev: Marked as regression because -21.644043043609884 (T) = (0 -48.72163261037674) / Math.Sqrt((8.11892158472017 / (34)) + (0.6095720265323639 / (2))) is less than -2.0322445093148245 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (34) + (2) - 2, .025) and -0.48704907541745585 = (32.763970884215254 - 48.72163261037674) / 32.763970884215254 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.

Docs

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

@radekdoulik
Copy link
Member

radekdoulik commented Apr 20, 2023

this is probably emscripten bump to 3.1.34 dotnet/runtime#83998
range: dotnet/runtime@7dbf6a5...223d152

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