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] Changes at 5/19/2022 4:33:45 PM #69724

Closed
performanceautofiler bot opened this issue May 24, 2022 · 10 comments
Closed

[Perf] Changes at 5/19/2022 4:33:45 PM #69724

performanceautofiler bot opened this issue May 24, 2022 · 10 comments
Assignees
@DrewScoggins
Labels
arch-x86 area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI runtime-coreclr specific to the CoreCLR runtime tenet-performance Performance related issue tenet-performance-benchmarks Issue from performance benchmark
Milestone
7.0.0

Comments

@performanceautofiler
Copy link

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in Microsoft.Extensions.Logging.FormattingOverhead

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TwoArguments_DefineMessage - Duration of single invocation 190.28 ns 208.55 ns 1.10 0.05 False
FourArguments_DefineMessage - Duration of single invocation 329.64 ns 357.77 ns 1.09 0.06 False

graph
graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'Microsoft.Extensions.Logging.FormattingOverhead*'

Payloads

Baseline
Compare

Histogram

Microsoft.Extensions.Logging.FormattingOverhead.TwoArguments_DefineMessage

Description of detection logic

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 208.54658660207673 > 200.16646858002863.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -14.42011962469898 (T) = (0 -204.24463804258718) / Math.Sqrt((20.513812963662776 / (32)) + (5.266395806044664 / (25))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (32) + (25) - 2, .025) and -0.0696989569425386 = (190.93655903560787 - 204.24463804258718) / 190.93655903560787 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.

```#### Microsoft.Extensions.Logging.FormattingOverhead.FourArguments_DefineMessage

```log

Description of detection logic

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 357.7671982836718 > 340.5513374173348.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -11.343891593974547 (T) = (0 -351.84479509828975) / Math.Sqrt((18.596433453626204 / (30)) + (93.36178514595002 / (25))) is less than -2.005745995316835 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (30) + (25) - 2, .025) and -0.07213075606830527 = (328.1733996593544 - 351.84479509828975) / 328.1733996593544 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
Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in Microsoft.Extensions.Logging.Formatting

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TwoArguments - Duration of single invocation 192.09 ns 207.68 ns 1.08 0.03 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'Microsoft.Extensions.Logging.Formatting*'

Payloads

Baseline
Compare

Histogram

Microsoft.Extensions.Logging.Formatting.TwoArguments

Description of detection logic

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 207.67838109501062 > 201.88087968295798.
IsChangePoint: Marked as a change because one of 4/16/2022 10:12:41 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -13.70789230698465 (T) = (0 -203.57779332427245) / Math.Sqrt((14.310321253963135 / (33)) + (10.281239535251894 / (24))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (24) - 2, .025) and -0.0666866011614706 = (190.85061451283354 - 203.57779332427245) / 190.85061451283354 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

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.Tests.Perf_String

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Format_MultipleArgs - Duration of single invocation 289.72 ns 334.64 ns 1.16 0.02 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Tests.Perf_String*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_String.Format_MultipleArgs

Description of detection logic

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 334.64170375597047 > 303.8066111818096.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -64.29305598726903 (T) = (0 -334.5725368743201) / Math.Sqrt((6.5107715575410205 / (33)) + (7.319639993940108 / (25))) is less than -2.0032407188469383 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (25) - 2, .025) and -0.15543695500207153 = (289.56364553332145 - 334.5725368743201) / 289.56364553332145 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

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.Globalization.Tests.StringEquality

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Compare_Same - Duration of single invocation 1.03 μs 1.30 μs 1.26 0.00 True

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Globalization.Tests.StringEquality*'

Payloads

Baseline
Compare

Histogram

System.Globalization.Tests.StringEquality.Compare_Same(Count: 1024, Options: (en-US, OrdinalIgnoreCase))

Description of detection logic

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 1.2981847889434683 > 1.082240769364866.
IsChangePoint: Marked as a change because one of 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -166.9768272494324 (T) = (0 -1297.272778122098) / Math.Sqrt((82.13182406233398 / (33)) + (0.8082664463318828 / (25))) is less than -2.0032407188469383 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (25) - 2, .025) and -0.25687242934330734 = (1032.1435555714268 - 1297.272778122098) / 1032.1435555714268 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 as regression because Edge Detector said so.

Docs

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

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.IO.Tests.Perf_StreamWriter

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
WriteFormat - Duration of single invocation 121.48 ns 152.53 ns 1.26 0.01 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.IO.Tests.Perf_StreamWriter*'

Payloads

Baseline
Compare

Histogram

System.IO.Tests.Perf_StreamWriter.WriteFormat

Description of detection logic

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 152.52610072861836 > 127.9248302742737.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -118.11814837328622 (T) = (0 -153.1034076811276) / Math.Sqrt((1.0526317590992857 / (33)) + (0.836670892446692 / (24))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (24) - 2, .025) and -0.24896454796812012 = (122.5842702502678 - 153.1034076811276) / 122.5842702502678 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
@performanceautofiler performanceautofiler bot added CoreClr untriaged New issue has not been triaged by the area owner labels May 24, 2022
@dotnet-issue-labeler
Copy link

I couldn't figure out the best area label to add to this issue. If you have write-permissions please help me learn by adding exactly one area label.

@kunalspathak kunalspathak transferred this issue from dotnet/perf-autofiling-issues May 24, 2022
@kunalspathak
Copy link
Member

@DrewScoggins - Seems we didn't have run for few days. Can we do catchup runs to narrow down the commit that caused the regression?

@kunalspathak kunalspathak added tenet-performance Performance related issue tenet-performance-benchmarks Issue from performance benchmark labels May 24, 2022
@kunalspathak kunalspathak removed their assignment May 26, 2022
@jeffschwMSFT
Copy link
Member

Assigning to Extensions label until we can get an updated commit range.

@ghost
Copy link

ghost commented May 26, 2022

Tagging subscribers to this area: @dotnet/area-extensions-logging
See info in area-owners.md if you want to be subscribed.

Issue Details

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in Microsoft.Extensions.Logging.FormattingOverhead

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TwoArguments_DefineMessage - Duration of single invocation 190.28 ns 208.55 ns 1.10 0.05 False
FourArguments_DefineMessage - Duration of single invocation 329.64 ns 357.77 ns 1.09 0.06 False

graph
graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'Microsoft.Extensions.Logging.FormattingOverhead*'

Payloads

Baseline
Compare

Histogram

Microsoft.Extensions.Logging.FormattingOverhead.TwoArguments_DefineMessage

Description of detection logic

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 208.54658660207673 > 200.16646858002863.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -14.42011962469898 (T) = (0 -204.24463804258718) / Math.Sqrt((20.513812963662776 / (32)) + (5.266395806044664 / (25))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (32) + (25) - 2, .025) and -0.0696989569425386 = (190.93655903560787 - 204.24463804258718) / 190.93655903560787 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.

```#### Microsoft.Extensions.Logging.FormattingOverhead.FourArguments_DefineMessage

```log

Description of detection logic

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 357.7671982836718 > 340.5513374173348.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -11.343891593974547 (T) = (0 -351.84479509828975) / Math.Sqrt((18.596433453626204 / (30)) + (93.36178514595002 / (25))) is less than -2.005745995316835 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (30) + (25) - 2, .025) and -0.07213075606830527 = (328.1733996593544 - 351.84479509828975) / 328.1733996593544 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
Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in Microsoft.Extensions.Logging.Formatting

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TwoArguments - Duration of single invocation 192.09 ns 207.68 ns 1.08 0.03 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'Microsoft.Extensions.Logging.Formatting*'

Payloads

Baseline
Compare

Histogram

Microsoft.Extensions.Logging.Formatting.TwoArguments

Description of detection logic

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 207.67838109501062 > 201.88087968295798.
IsChangePoint: Marked as a change because one of 4/16/2022 10:12:41 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -13.70789230698465 (T) = (0 -203.57779332427245) / Math.Sqrt((14.310321253963135 / (33)) + (10.281239535251894 / (24))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (24) - 2, .025) and -0.0666866011614706 = (190.85061451283354 - 203.57779332427245) / 190.85061451283354 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

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.Tests.Perf_String

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Format_MultipleArgs - Duration of single invocation 289.72 ns 334.64 ns 1.16 0.02 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Tests.Perf_String*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_String.Format_MultipleArgs

Description of detection logic

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 334.64170375597047 > 303.8066111818096.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -64.29305598726903 (T) = (0 -334.5725368743201) / Math.Sqrt((6.5107715575410205 / (33)) + (7.319639993940108 / (25))) is less than -2.0032407188469383 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (25) - 2, .025) and -0.15543695500207153 = (289.56364553332145 - 334.5725368743201) / 289.56364553332145 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

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.Globalization.Tests.StringEquality

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Compare_Same - Duration of single invocation 1.03 μs 1.30 μs 1.26 0.00 True

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Globalization.Tests.StringEquality*'

Payloads

Baseline
Compare

Histogram

System.Globalization.Tests.StringEquality.Compare_Same(Count: 1024, Options: (en-US, OrdinalIgnoreCase))

Description of detection logic

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 1.2981847889434683 > 1.082240769364866.
IsChangePoint: Marked as a change because one of 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -166.9768272494324 (T) = (0 -1297.272778122098) / Math.Sqrt((82.13182406233398 / (33)) + (0.8082664463318828 / (25))) is less than -2.0032407188469383 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (25) - 2, .025) and -0.25687242934330734 = (1032.1435555714268 - 1297.272778122098) / 1032.1435555714268 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 as regression because Edge Detector said so.

Docs

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

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.IO.Tests.Perf_StreamWriter

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
WriteFormat - Duration of single invocation 121.48 ns 152.53 ns 1.26 0.01 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.IO.Tests.Perf_StreamWriter*'

Payloads

Baseline
Compare

Histogram

System.IO.Tests.Perf_StreamWriter.WriteFormat

Description of detection logic

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 152.52610072861836 > 127.9248302742737.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -118.11814837328622 (T) = (0 -153.1034076811276) / Math.Sqrt((1.0526317590992857 / (33)) + (0.836670892446692 / (24))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (24) - 2, .025) and -0.24896454796812012 = (122.5842702502678 - 153.1034076811276) / 122.5842702502678 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

Author: performanceautofiler[bot]
Assignees: DrewScoggins
Labels:

tenet-performance, tenet-performance-benchmarks, untriaged, area-Extensions-Logging, refs/heads/main, RunKind=micro, Windows 10.0.18362, Regression, CoreClr, x86
Milestone: -

@maryamariyan maryamariyan removed the untriaged New issue has not been triaged by the area owner label May 31, 2022
@maryamariyan maryamariyan added this to the 7.0.0 milestone May 31, 2022
@dakersnar
Copy link
Contributor

More regression data for System.Globalization.Tests.StringEquality.Compare_Same(Count: 1024, Options: (en-US, OrdinalIgnoreCase)) from the perf report:

System.Globalization.Tests.StringEquality.Compare_Same(Count: 1024, Options: (en-US, OrdinalIgnoreCase))

Result Ratio Operating System Bit Processor Name
Same 1.00 debian 11 Arm64 Unknown processor
Slower 0.87 ubuntu 18.04 Arm64 Unknown processor
Same 1.00 ubuntu 20.04 Arm64 Unknown processor
Same 1.00 Windows 11 Arm64 Microsoft SQ1 3.0 GHz
Slower 0.32 macOS Monterey 12.3 Arm64 Apple M1 Max
Slower 0.78 Windows 10 X64 Intel Core i7-6700 CPU 3.40GHz (Skylake)
Slower 0.74 Windows 10 X64 Intel Core i7-8650U CPU 1.90GHz (Kaby Lake R)
Slower 0.72 Windows 10 X64 Intel Core i9-10900K CPU 3.70GHz
Slower 0.67 Windows 11 X64 AMD Ryzen 9 5900X
Slower 0.66 Windows 11 X64 AMD Ryzen 9 5950X
Slower 0.77 Windows 11 X64 Intel Core i7-8700 CPU 3.20GHz (Coffee Lake)
Same 0.96 Windows 11 X64 11th Gen Intel Core i9-11900H 2.50GHz
Slower 0.79 Windows 11 X64 Intel Core i9-9900T CPU 2.10GHz
Same 1.00 ubuntu 18.04 X64 Intel Xeon CPU E5530 2.40GHz
Slower 0.85 ubuntu 18.04 X64 Intel Core i7-2720QM CPU 2.20GHz (Sandy Bridge)
Slower 0.74 ubuntu 20.04 X64 Intel Core i7-8700 CPU 3.20GHz (Coffee Lake)
Slower 0.87 Windows 10 X86 Intel Core i7-6700 CPU 3.40GHz (Skylake)
Same 0.99 macOS Big Sur 11.6.6 X64 Intel Core i5-4278U CPU 2.60GHz (Haswell)

@dakersnar
Copy link
Contributor

More regression data for System.Tests.Perf_String.Format_MultipleArgs from the perf report:

System.Tests.Perf_String.Format_MultipleArgs

Result Ratio Operating System Bit Processor Name
Same 0.97 debian 11 Arm64 Unknown processor
Same 0.92 ubuntu 18.04 Arm64 Unknown processor
Same 0.92 ubuntu 20.04 Arm64 Unknown processor
Slower 0.86 Windows 11 Arm64 Microsoft SQ1 3.0 GHz
Slower 0.88 macOS Monterey 12.3 Arm64 Apple M1 Max
Slower 0.81 Windows 10 X64 Intel Core i7-6700 CPU 3.40GHz (Skylake)
Same 0.93 Windows 10 X64 Intel Core i7-8650U CPU 1.90GHz (Kaby Lake R)
Slower 0.77 Windows 10 X64 Intel Core i9-10900K CPU 3.70GHz
Slower 0.88 Windows 11 X64 AMD Ryzen 9 5900X
Slower 0.87 Windows 11 X64 AMD Ryzen 9 5950X
Slower 0.85 Windows 11 X64 Intel Core i7-8700 CPU 3.20GHz (Coffee Lake)
Same 0.89 Windows 11 X64 11th Gen Intel Core i9-11900H 2.50GHz
Slower 0.86 Windows 11 X64 Intel Core i9-9900T CPU 2.10GHz
Same 0.94 ubuntu 18.04 X64 Intel Xeon CPU E5530 2.40GHz
Slower 0.84 ubuntu 18.04 X64 Intel Core i7-2720QM CPU 2.20GHz (Sandy Bridge)
Slower 0.79 ubuntu 20.04 X64 Intel Core i7-8700 CPU 3.20GHz (Coffee Lake)
Same 0.89 Windows 10 X86 Intel Core i7-6700 CPU 3.40GHz (Skylake)
Slower 0.89 macOS Big Sur 11.6.6 X64 Intel Core i5-4278U CPU 2.60GHz (Haswell)

@dakersnar dakersnar mentioned this issue Jun 10, 2022
Closed
14 tasks
@tarekgh
Copy link
Member

tarekgh commented Aug 3, 2022

Looking at the list of commits 6e4eba8...15dda0e I am not seeing any commit for logging nor globalization which can cause any regression.

@AndyAyersMS do you think 41491a3 can cause such regression?

@tarekgh tarekgh added area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI and removed area-Extensions-Logging labels Aug 3, 2022
@ghost
Copy link

ghost commented Aug 3, 2022

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

Issue Details

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in Microsoft.Extensions.Logging.FormattingOverhead

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TwoArguments_DefineMessage - Duration of single invocation 190.28 ns 208.55 ns 1.10 0.05 False
FourArguments_DefineMessage - Duration of single invocation 329.64 ns 357.77 ns 1.09 0.06 False

graph
graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'Microsoft.Extensions.Logging.FormattingOverhead*'

Payloads

Baseline
Compare

Histogram

Microsoft.Extensions.Logging.FormattingOverhead.TwoArguments_DefineMessage

Description of detection logic

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 208.54658660207673 > 200.16646858002863.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -14.42011962469898 (T) = (0 -204.24463804258718) / Math.Sqrt((20.513812963662776 / (32)) + (5.266395806044664 / (25))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (32) + (25) - 2, .025) and -0.0696989569425386 = (190.93655903560787 - 204.24463804258718) / 190.93655903560787 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.

```#### Microsoft.Extensions.Logging.FormattingOverhead.FourArguments_DefineMessage

```log

Description of detection logic

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 357.7671982836718 > 340.5513374173348.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -11.343891593974547 (T) = (0 -351.84479509828975) / Math.Sqrt((18.596433453626204 / (30)) + (93.36178514595002 / (25))) is less than -2.005745995316835 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (30) + (25) - 2, .025) and -0.07213075606830527 = (328.1733996593544 - 351.84479509828975) / 328.1733996593544 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
Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in Microsoft.Extensions.Logging.Formatting

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
TwoArguments - Duration of single invocation 192.09 ns 207.68 ns 1.08 0.03 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'Microsoft.Extensions.Logging.Formatting*'

Payloads

Baseline
Compare

Histogram

Microsoft.Extensions.Logging.Formatting.TwoArguments

Description of detection logic

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 207.67838109501062 > 201.88087968295798.
IsChangePoint: Marked as a change because one of 4/16/2022 10:12:41 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -13.70789230698465 (T) = (0 -203.57779332427245) / Math.Sqrt((14.310321253963135 / (33)) + (10.281239535251894 / (24))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (24) - 2, .025) and -0.0666866011614706 = (190.85061451283354 - 203.57779332427245) / 190.85061451283354 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

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.Tests.Perf_String

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Format_MultipleArgs - Duration of single invocation 289.72 ns 334.64 ns 1.16 0.02 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Tests.Perf_String*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_String.Format_MultipleArgs

Description of detection logic

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 334.64170375597047 > 303.8066111818096.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -64.29305598726903 (T) = (0 -334.5725368743201) / Math.Sqrt((6.5107715575410205 / (33)) + (7.319639993940108 / (25))) is less than -2.0032407188469383 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (25) - 2, .025) and -0.15543695500207153 = (289.56364553332145 - 334.5725368743201) / 289.56364553332145 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

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.Globalization.Tests.StringEquality

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Compare_Same - Duration of single invocation 1.03 μs 1.30 μs 1.26 0.00 True

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Globalization.Tests.StringEquality*'

Payloads

Baseline
Compare

Histogram

System.Globalization.Tests.StringEquality.Compare_Same(Count: 1024, Options: (en-US, OrdinalIgnoreCase))

Description of detection logic

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 1.2981847889434683 > 1.082240769364866.
IsChangePoint: Marked as a change because one of 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -166.9768272494324 (T) = (0 -1297.272778122098) / Math.Sqrt((82.13182406233398 / (33)) + (0.8082664463318828 / (25))) is less than -2.0032407188469383 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (25) - 2, .025) and -0.25687242934330734 = (1032.1435555714268 - 1297.272778122098) / 1032.1435555714268 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 as regression because Edge Detector said so.

Docs

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

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline 6e4eba8c98f4e05ece31a883dc23076a78f7c7b9
Compare 15dda0efe52c2aaa7a4e783f443eefb558529248
Diff Diff

Regressions in System.IO.Tests.Perf_StreamWriter

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
WriteFormat - Duration of single invocation 121.48 ns 152.53 ns 1.26 0.01 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.IO.Tests.Perf_StreamWriter*'

Payloads

Baseline
Compare

Histogram

System.IO.Tests.Perf_StreamWriter.WriteFormat

Description of detection logic

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 152.52610072861836 > 127.9248302742737.
IsChangePoint: Marked as a change because one of 4/16/2022 5:36:25 PM, 5/18/2022 10:04:15 AM, 5/24/2022 7:58:36 AM falls between 5/15/2022 6:57:22 PM and 5/24/2022 7:58:36 AM.
IsRegressionStdDev: Marked as regression because -118.11814837328622 (T) = (0 -153.1034076811276) / Math.Sqrt((1.0526317590992857 / (33)) + (0.836670892446692 / (24))) is less than -2.0040447832881556 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (24) - 2, .025) and -0.24896454796812012 = (122.5842702502678 - 153.1034076811276) / 122.5842702502678 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

Author: performanceautofiler[bot]
Assignees: DrewScoggins
Labels:

tenet-performance, tenet-performance-benchmarks, area-CodeGen-coreclr, refs/heads/main, RunKind=micro, Windows 10.0.18362, Regression, CoreClr, x86
Milestone: 7.0.0

@AndyAyersMS
Copy link
Member

@AndyAyersMS do you think 41491a3 can cause such regression?

No, it was a bug fix for a very rare case.

Seems like whatever the cause of these regressions, most now are faster now -- perhaps we can just close this.

newplot - 2022-08-02T195059 943
newplot - 2022-08-02T195202 485
newplot - 2022-08-02T195249 957

@tarekgh
Copy link
Member

tarekgh commented Aug 3, 2022

Per @AndyAyersMS reply, I am closing this issue for now as the regression is not there anymore. Thanks @AndyAyersMS!

@tarekgh tarekgh closed this as completed Aug 3, 2022
@ghost ghost locked as resolved and limited conversation to collaborators Sep 2, 2022
@jeffhandley jeffhandley added runtime-coreclr specific to the CoreCLR runtime arch-x86 and removed CoreClr labels Dec 28, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees

@DrewScoggins DrewScoggins

Labels
arch-x86 area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI runtime-coreclr specific to the CoreCLR runtime tenet-performance Performance related issue tenet-performance-benchmarks Issue from performance benchmark
Projects
None yet
Milestone
7.0.0
Development

No branches or pull requests
8 participants
@jeffhandley @maryamariyan @DrewScoggins @jeffschwMSFT @AndyAyersMS @tarekgh @kunalspathak @dakersnar