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: 9 Regressions on 4/17/2023 1:30:24 AM #16927

Closed
performanceautofiler bot opened this issue Apr 25, 2023 · 1 comment
Closed

[Perf] Linux/x64: 9 Regressions on 4/17/2023 1:30:24 AM #16927

performanceautofiler bot opened this issue Apr 25, 2023 · 1 comment
Assignees
@kunalspathak
Labels
arch-x64 branch-refs/heads/main kind-micro os-linux perf-regression runtime-coreclr untriaged

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Apr 25, 2023 •
edited
Loading

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 7526a4c6a0a095670f1135d983a349b4e867f2e2
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Text.Json.Serialization.Tests.WriteJson<SimpleStructWithProperties>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
SerializeToWriter - Duration of single invocation 81.37 ns 86.81 ns 1.07 0.02 False
SerializeToUtf8Bytes - Duration of single invocation 146.75 ns 157.48 ns 1.07 0.14 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.Text.Json.Serialization.Tests.WriteJson&lt;SimpleStructWithProperties&gt;*'

Payloads

Baseline
Compare

Histogram

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

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 86.81417176693381 > 86.07156512671189.
IsChangePoint: Marked as a change because one of 2/24/2023 8:06:06 PM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -13.884687250814741 (T) = (0 -88.70042079226445) / Math.Sqrt((1.0381009518636748 / (12)) + (4.067164806659238 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.07454913027940242 = (82.54664053303986 - 88.70042079226445) / 82.54664053303986 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.Text.Json.Serialization.Tests.WriteJson<SimpleStructWithProperties>.SerializeToUtf8Bytes(Mode: SourceGen)

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 157.4820417635919 > 155.43608158463329.
IsChangePoint: Marked as a change because one of 4/17/2023 6:40:18 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -7.5545474001054975 (T) = (0 -158.23937663686988) / Math.Sqrt((24.08518519351257 / (16)) + (7.953368762542879 / (33))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (16) + (33) - 2, .025) and -0.06733790990019284 = (148.25611942488476 - 158.23937663686988) / 148.25611942488476 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 7526a4c6a0a095670f1135d983a349b4e867f2e2
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in PerfLabTests.BlockCopyPerf

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
CallBlockCopy - Duration of single invocation 12.90 ns 16.79 ns 1.30 0.31 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 'PerfLabTests.BlockCopyPerf*'

Payloads

Baseline
Compare

Histogram

PerfLabTests.BlockCopyPerf.CallBlockCopy(numElements: 1000)

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 16.79261134742772 > 13.547404071096906.
IsChangePoint: Marked as a change because one of 3/22/2023 10:50:22 PM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -4.8431402758149185 (T) = (0 -17.124053150466644) / Math.Sqrt((5.79962482215307 / (12)) + (2.8829499756852837 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.268839908959585 = (13.495834288904037 - 17.124053150466644) / 13.495834288904037 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 7526a4c6a0a095670f1135d983a349b4e867f2e2
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_Int32

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToStringHex - Duration of single invocation 16.04 ns 17.63 ns 1.10 0.17 False
ToStringHex - Duration of single invocation 18.72 ns 21.99 ns 1.18 0.12 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_Int32*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Int32.ToStringHex(value: 12345)

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 17.630579738188164 > 16.849269058228746.
IsChangePoint: Marked as a change because one of 2/9/2023 4:27:10 AM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -17.872191079918526 (T) = (0 -17.939966077851988) / Math.Sqrt((0.13629602003049193 / (12)) + (0.06820872578855902 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.12925909195794175 = (15.886492484862053 - 17.939966077851988) / 15.886492484862053 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_Int32.ToStringHex(value: -2147483648)

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 21.992324266046 > 19.59691986891323.
IsChangePoint: Marked as a change because one of 2/9/2023 4:27:10 AM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -37.35518803374646 (T) = (0 -22.4964699514032) / Math.Sqrt((0.09484723603291974 / (12)) + (0.11780291520455705 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.211897073168776 = (18.563020284042064 - 22.4964699514032) / 18.563020284042064 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 7526a4c6a0a095670f1135d983a349b4e867f2e2
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_Int64

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToString - Duration of single invocation 29.86 ns 31.85 ns 1.07 0.21 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_Int64*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Int64.ToString(value: -9223372036854775808)

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 31.851427820133527 > 31.542823705377835.
IsChangePoint: Marked as a change because one of 2/9/2023 12:40:40 PM, 4/6/2023 1:55:21 PM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -11.371874324634387 (T) = (0 -31.886220031841976) / Math.Sqrt((0.4794897695332556 / (12)) + (0.2541020501841517 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.08362735206482456 = (29.425447752941626 - 31.886220031841976) / 29.425447752941626 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 7526a4c6a0a095670f1135d983a349b4e867f2e2
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Text.RegularExpressions.Tests.Perf_Regex_Industry_Leipzig

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Count - Duration of single invocation 55.30 ms 61.90 ms 1.12 0.20 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.Text.RegularExpressions.Tests.Perf_Regex_Industry_Leipzig*'

Payloads

Baseline
Compare

Histogram

System.Text.RegularExpressions.Tests.Perf_Regex_Industry_Leipzig.Count(Pattern: ".{0,2}(Tom|Sawyer|Huckleberry|Finn)", Options: NonBacktracking)

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 61.89906033333334 > 58.082495793750006.
IsChangePoint: Marked as a change because one of 2/24/2023 8:06:06 PM, 3/28/2023 3:45:48 AM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -11.001083984326334 (T) = (0 -61987372.82716068) / Math.Sqrt((3623179246533.4136 / (12)) + (9682002190.455929 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.1081076988315377 = (55939844.91987943 - 61987372.82716068) / 55939844.91987943 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 7526a4c6a0a095670f1135d983a349b4e867f2e2
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_DateTime

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToString - Duration of single invocation 26.11 ns 29.75 ns 1.14 0.04 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

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

Payloads

Baseline
Compare

Histogram

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

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 29.74591212232118 > 26.808858710801385.
IsChangePoint: Marked as a change because one of 4/9/2023 4:09:45 AM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -19.489524121035636 (T) = (0 -29.320633078849944) / Math.Sqrt((0.37218615749084377 / (12)) + (0.4584672463657675 / (36))) is less than -2.0128955989180297 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (36) - 2, .025) and -0.1614858136615133 = (25.24407335326674 - 29.320633078849944) / 25.24407335326674 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 7526a4c6a0a095670f1135d983a349b4e867f2e2
Compare c1b7a9feb6f3b4d9ca27dc4f74d8260e4edb73e8
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

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

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
GetHashCodeBenchmark - Duration of single invocation 7.40 ns 8.55 ns 1.16 0.10 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.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int64&gt;*'

Payloads

Baseline
Compare

Histogram

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

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 8.549270613609094 > 7.779017040323971.
IsChangePoint: Marked as a change because one of 2/9/2023 11:32:54 PM, 4/12/2023 8:39:35 PM, 4/25/2023 3:46:39 AM falls between 4/12/2023 8:39:35 PM and 4/25/2023 3:46:39 AM.
IsRegressionStdDev: Marked as regression because -16.04766142993529 (T) = (0 -8.390764490712163) / Math.Sqrt((0.0006230187198144381 / (12)) + (0.13554001586866155 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.13195582704901435 = (7.412625378312433 - 8.390764490712163) / 7.412625378312433 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
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@kunalspathak kunalspathak

Labels
arch-x64 branch-refs/heads/main kind-micro os-linux perf-regression runtime-coreclr untriaged
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@kunalspathak