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] Changes at 6/27/2022 5:19:23 PM #6523

Closed
performanceautofiler bot opened this issue Jul 5, 2022 · 3 comments
Closed

[Perf] Changes at 6/27/2022 5:19:23 PM #6523

performanceautofiler bot opened this issue Jul 5, 2022 · 3 comments
Assignees
Labels
arch-x86 branch-refs/heads/main kind-micro os-windows perf-regression PGO Applied if there were any profile guided optimization updates in the observed interval. runtime-coreclr untriaged

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented Jul 5, 2022

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Collections.TryGetValueTrue<String, String>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
IDictionary - Duration of single invocation 16.96 μs 20.09 μs 1.18 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.Collections.TryGetValueTrue&lt;String, String&gt;*'

Payloads

Baseline
Compare

Histogram

System.Collections.TryGetValueTrue<String, String>.IDictionary(Size: 512)


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 20.090707314457315 > 17.93734161820127.
IsChangePoint: Marked as a change because one of 5/11/2022 9:09:01 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -59.074240363476164 (T) = (0 -20031.31436306825) / Math.Sqrt((25010.96931407557 / (13)) + (21549.537621726875 / (39))) is less than -2.0085591120996775 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (13) + (39) - 2, .025) and -0.17200216271802707 = (17091.533616809204 - 20031.31436306825) / 17091.533616809204 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

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
ToStringL - Duration of single invocation 104.98 ns 135.55 ns 1.29 0.02 False
TryFormatL - Duration of single invocation 89.63 ns 119.49 ns 1.33 0.01 False

graph
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_Version*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Version.ToStringL


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 135.54983781036518 > 110.53936462885197.
IsChangePoint: Marked as a change because one of 5/5/2022 12:13:02 AM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -153.881708837624 (T) = (0 -135.7324307850285) / Math.Sqrt((0.4072879237403935 / (12)) + (0.12832779232885416 / (35))) is less than -2.0141033888794695 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (35) - 2, .025) and -0.28181494737024415 = (105.89081603666386 - 135.7324307850285) / 105.89081603666386 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_Version.TryFormatL

```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 119.49324122430951 > 94.12936988118697.
IsChangePoint: Marked as a change because one of 5/5/2022 12:13:02 AM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -146.43123492479535 (T) = (0 -119.83821822812234) / Math.Sqrt((0.02957987888263038 / (10)) + (1.4991548678760405 / (38))) is less than -2.0128955989180297 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (10) + (38) - 2, .025) and -0.33624486207382354 = (89.68282807249562 - 119.83821822812234) / 89.68282807249562 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Text.Perf_Utf8Encoding

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
GetBytes - Duration of single invocation 214.80 μs 229.12 μs 1.07 0.00 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Text.Perf_Utf8Encoding*'

Payloads

Baseline
Compare

Histogram

System.Text.Perf_Utf8Encoding.GetBytes(Input: Greek)


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 229.12461180124222 > 226.52214192708334.
IsChangePoint: Marked as a change because one of 5/16/2022 5:45:53 PM, 5/31/2022 6:06:22 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -19.905073583606114 (T) = (0 -227986.3350365568) / Math.Sqrt((2815645.962568115 / (12)) + (3553703.014274096 / (39))) is less than -2.0095752371279447 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (39) - 2, .025) and -0.05244470392852297 = (216625.4760801576 - 227986.3350365568) / 216625.4760801576 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

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
ToString - Duration of single invocation 11.55 ns 13.16 ns 1.14 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 'System.Tests.Perf_Int16*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Int16.ToString(value: 32767)


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 13.157181720360864 > 12.164430364912972.
IsChangePoint: Marked as a change because one of 5/9/2022 5:23:46 PM, 5/30/2022 12:57:05 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -83.03108658945924 (T) = (0 -13.204599871367074) / Math.Sqrt((0.002640416036380294 / (11)) + (0.0046689873228349 / (35))) is less than -2.0153675744421933 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (11) + (35) - 2, .025) and -0.13832137827543398 = (11.600063148574218 - 13.204599871367074) / 11.600063148574218 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Tests.Perf_SByte

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
ToString - Duration of single invocation 8.85 ns 10.01 ns 1.13 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 'System.Tests.Perf_SByte*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_SByte.ToString(value: 127)


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 10.005480265327012 > 9.292779266369914.
IsChangePoint: Marked as a change because one of 5/9/2022 5:23:46 PM, 5/30/2022 12:57:05 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -37.77203957474321 (T) = (0 -10.04146170055147) / Math.Sqrt((0.0011242657975215722 / (12)) + (0.03431124061614244 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.13661698276408352 = (8.834516686643312 - 10.04146170055147) / 8.834516686643312 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Memory.Slice<Byte>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
MemoryStartSpan - Duration of single invocation 2.77 ns 4.96 ns 1.79 0.07 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Memory.Slice&lt;Byte&gt;*'

Payloads

Baseline
Compare

Histogram

System.Memory.Slice<Byte>.MemoryStartSpan


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 4.955379807663401 > 2.916418158257601.
IsChangePoint: Marked as a change because one of 5/20/2022 10:18:01 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -113.52736863424414 (T) = (0 -4.998623591435183) / Math.Sqrt((0.004025279597783472 / (14)) + (0.0028156843704145542 / (32))) is less than -2.0153675744421933 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (14) + (32) - 2, .025) and -0.7860642035766832 = (2.798680798498278 - 4.998623591435183) / 2.798680798498278 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Tests.Perf_Uri

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Ctor - Duration of single invocation 107.94 ns 114.50 ns 1.06 0.04 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_Uri*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Uri.Ctor


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 114.50181639320365 > 113.51764074916692.
IsChangePoint: Marked as a change because one of 6/2/2022 5:58:32 PM, 6/7/2022 6:20:49 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -23.93097709507501 (T) = (0 -115.50484054067753) / Math.Sqrt((0.4671489250103736 / (12)) + (1.4908296880047838 / (37))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (37) - 2, .025) and -0.061926567262963333 = (108.76914101356618 - 115.50484054067753) / 108.76914101356618 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in V8.Crypto.Support

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Bench - Duration of single invocation 7.36 ms 7.92 ms 1.08 0.00 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'V8.Crypto.Support*'

Payloads

Baseline
Compare

Histogram

V8.Crypto.Support.Bench


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 7.918412258064516 > 7.616188485294117.
IsChangePoint: Marked as a change because one of 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -39.74818310971154 (T) = (0 -7922488.893002421) / Math.Sqrt((3173858083.759653 / (14)) + (1243256974.1608372 / (32))) is less than -2.0153675744421933 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (14) + (32) - 2, .025) and -0.0890382862243076 = (7274756.997267438 - 7922488.893002421) / 7274756.997267438 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
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_Upper - Duration of single invocation 1.49 μs 1.65 μs 1.11 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.Globalization.Tests.StringEquality*'

Payloads

Baseline
Compare

Histogram

System.Globalization.Tests.StringEquality.Compare_Same_Upper(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 1.6529420619960205 > 1.5723748063163028.
IsChangePoint: Marked as a change because one of 5/18/2022 10:04:15 AM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -63.35475693818205 (T) = (0 -1659.4830665354345) / Math.Sqrt((50.96137322411445 / (11)) + (74.70140068005917 / (39))) is less than -2.010634757623041 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (11) + (39) - 2, .025) and -0.10827199515818 = (1497.360822781218 - 1659.4830665354345) / 1497.360822781218 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 988f570dc4ddd79a166ca8cbc184b2938ed2fefc
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Collections.AddGivenSize<String>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
HashSet - Duration of single invocation 17.29 μs 19.08 μs 1.10 0.26 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Collections.AddGivenSize&lt;String&gt;*'

Payloads

Baseline
Compare

Histogram

System.Collections.AddGivenSize<String>.HashSet(Size: 512)


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 19.07746958158263 > 18.95716508510954.
IsChangePoint: Marked as a change because one of 5/30/2022 5:18:40 PM, 6/26/2022 6:11:06 PM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -4.604384194296194 (T) = (0 -19553.32526464886) / Math.Sqrt((2210995.7559413156 / (11)) + (365820.0912324677 / (40))) is less than -2.0095752371279447 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (11) + (40) - 2, .025) and -0.12100962720171221 = (17442.60244531377 - 19553.32526464886) / 17442.60244531377 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Memory.Span<Char>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
StartsWith - Duration of single invocation 13.76 ns 15.81 ns 1.15 0.35 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Memory.Span&lt;Char&gt;*'

Payloads

Baseline
Compare

Histogram

System.Memory.Span<Char>.StartsWith(Size: 512)


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 15.810275924189785 > 14.447732852274873.
IsChangePoint: Marked as a change because one of 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -11.913118468021302 (T) = (0 -15.705075972465302) / Math.Sqrt((0.05411415853397115 / (13)) + (0.7912733446163401 / (39))) is less than -2.0085591120996775 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (13) + (39) - 2, .025) and -0.13457804478215954 = (13.84221741703163 - 15.705075972465302) / 13.84221741703163 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in PerfLabTests.LowLevelPerf

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
IntegerFormatting - Duration of single invocation 2.19 ms 2.31 ms 1.05 0.00 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

Histogram

PerfLabTests.LowLevelPerf.IntegerFormatting


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 2.3073209226190476 > 2.3054141785714286.
IsChangePoint: Marked as a change because one of 5/9/2022 11:56:12 AM, 5/24/2022 3:42:52 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -89.79026463182953 (T) = (0 -2310201.1147459988) / Math.Sqrt((12144260.818079213 / (12)) + (24338044.040397197 / (38))) is less than -2.010634757623041 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (38) - 2, .025) and -0.052590741292198526 = (2194776.2070469214 - 2310201.1147459988) / 2194776.2070469214 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Linq.Tests.Perf_Enumerable

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
OrderBy - Duration of single invocation 4.74 μs 5.05 μs 1.07 0.00 False
Where - Duration of single invocation 662.77 ns 753.60 ns 1.14 0.04 False
Reverse - Duration of single invocation 1.25 μs 1.42 μs 1.14 0.03 False
Average - Duration of single invocation 325.26 ns 506.08 ns 1.56 0.06 False

graph
graph
graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

Histogram

System.Linq.Tests.Perf_Enumerable.OrderBy(input: IEnumerable)


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 5.049938072159396 > 4.978843327627318.
IsChangePoint: Marked as a change because one of 5/30/2022 12:57:05 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -27.747707714294993 (T) = (0 -4991.925070293302) / Math.Sqrt((73.48158756226002 / (12)) + (2408.2148849700266 / (33))) is less than -2.016692199226234 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (33) - 2, .025) and -0.05200740665488232 = (4745.142513935679 - 4991.925070293302) / 4745.142513935679 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.Linq.Tests.Perf_Enumerable.Where(input: Array)

```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 753.5982993893812 > 695.917201817853.
IsChangePoint: Marked as a change because one of 5/9/2022 11:56:12 AM, 5/11/2022 9:09:01 PM, 5/28/2022 5:31:50 AM, 5/31/2022 9:21:33 PM, 6/6/2022 10:51:35 AM, 6/10/2022 11:27:32 AM, 6/22/2022 6:53:40 PM, 7/2/2022 7:35:06 PM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -52.421534321015656 (T) = (0 -753.1897412534973) / Math.Sqrt((116.04701134725855 / (42)) + (0.1684413667748993 / (7))) is less than -2.011740513728388 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (42) + (7) - 2, .025) and -0.13146896752676684 = (665.6742366517262 - 753.1897412534973) / 665.6742366517262 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.Linq.Tests.Perf_Enumerable.Reverse(input: IEnumerable)

```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 1.4232582346690956 > 1.3145192928032448.
IsChangePoint: Marked as a change because one of 5/9/2022 11:56:12 AM, 5/11/2022 9:09:01 PM, 5/28/2022 5:31:50 AM, 5/31/2022 9:21:33 PM, 6/6/2022 10:51:35 AM, 6/10/2022 11:27:32 AM, 6/22/2022 6:53:40 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -47.52685391356721 (T) = (0 -1389.9560684086118) / Math.Sqrt((1.146467006637895 / (13)) + (327.1261725808977 / (39))) is less than -2.0085591120996775 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (13) + (39) - 2, .025) and -0.11055392100292837 = (1251.5880968241133 - 1389.9560684086118) / 1251.5880968241133 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.Linq.Tests.Perf_Enumerable.Average(input: IEnumerable)

```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 506.0824099031646 > 341.62687882728505.
IsChangePoint: Marked as a change because one of 6/22/2022 10:05:16 PM, 6/30/2022 12:21:07 PM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -39.914342702671505 (T) = (0 -488.1171887173606) / Math.Sqrt((262.1200583475591 / (33)) + (190.52434709446834 / (19))) is less than -2.0085591120996775 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (33) + (19) - 2, .025) and -0.5305644461341974 = (318.9131891507189 - 488.1171887173606) / 318.9131891507189 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 PGO Applied if there were any profile guided optimization updates in the observed interval. untriaged labels Jul 5, 2022
@performanceautofiler
Copy link
Author

performanceautofiler bot commented Jul 5, 2022

Run Information

Architecture x86
OS Windows 10.0.18362
Baseline c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
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
WritePartialCharArray - Duration of single invocation 116.48 ms 136.87 ms 1.18 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.WritePartialCharArray(writeLength: 2)


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 136.87204666666665 > 122.26970874999999.
IsChangePoint: Marked as a change because one of 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -26.213038188243708 (T) = (0 -135760321.14239925) / Math.Sqrt((12550549490.410683 / (14)) + (19450525566823.035 / (36))) is less than -2.010634757623041 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (14) + (36) - 2, .025) and -0.165559172950046 = (116476558.45630561 - 135760321.14239925) / 116476558.45630561 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Collections.ContainsKeyTrue<String, String>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Dictionary - Duration of single invocation 14.35 μs 16.38 μs 1.14 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.Collections.ContainsKeyTrue&lt;String, String&gt;*'

Payloads

Baseline
Compare

Histogram

System.Collections.ContainsKeyTrue<String, String>.Dictionary(Size: 512)


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 16.375490945850913 > 15.120076588198735.
IsChangePoint: Marked as a change because one of 5/23/2022 7:50:08 PM, 5/24/2022 7:55:26 PM, 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -31.0893172076481 (T) = (0 -16384.287844111735) / Math.Sqrt((38497.961138846964 / (12)) + (29329.413741717657 / (35))) is less than -2.0141033888794695 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (35) - 2, .025) and -0.13726725929150557 = (14406.71725159732 - 16384.287844111735) / 14406.71725159732 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Tests.Perf_Random

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
Next_int_int_unseeded - Duration of single invocation 14.88 ns 15.89 ns 1.07 0.06 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_Random*'

Payloads

Baseline
Compare

Histogram

System.Tests.Perf_Random.Next_int_int_unseeded


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 15.886102000275416 > 15.628743323583105.
IsChangePoint: Marked as a change because one of 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -4.418622274209678 (T) = (0 -15.992361377833385) / Math.Sqrt((0.4876253604347575 / (13)) + (0.17572868605105255 / (42))) is less than -2.005745995316835 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (13) + (42) - 2, .025) and -0.059790080799475606 = (15.090121777483708 - 15.992361377833385) / 15.090121777483708 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 c9f80fdde29c71ac87971383c77a1a5e665fadb9
Compare 601ef262b5eca9a48a6207d4aa7f304c0827fdc8
Diff Diff

Regressions in System.Globalization.Tests.StringSearch

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
IsSuffix_DifferentLastChar - Duration of single invocation 16.97 ns 18.18 ns 1.07 0.27 False

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.StringSearch*'

Payloads

Baseline
Compare

Histogram

System.Globalization.Tests.StringSearch.IsSuffix_DifferentLastChar(Options: (en-US, Ordinal, False))


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 18.184209872511996 > 17.8254971943954.
IsChangePoint: Marked as a change because one of 6/27/2022 8:54:23 AM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -3.572033217019384 (T) = (0 -18.718906552903665) / Math.Sqrt((1.0012866006678138 / (12)) + (1.7069714870762747 / (38))) is less than -2.010634757623041 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (12) + (38) - 2, .025) and -0.07338491513263214 = (17.439136966621778 - 18.718906552903665) / 17.439136966621778 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 988f570dc4ddd79a166ca8cbc184b2938ed2fefc
Compare 844b0999fde67cdbaa7acb0f398ea1b066a2b8c7
Diff Diff

Regressions in System.Collections.CreateAddAndRemove<String>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
HashSet - Duration of single invocation 41.42 μs 45.19 μs 1.09 0.17 False

graph
Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Collections.CreateAddAndRemove&lt;String&gt;*'

Payloads

Baseline
Compare

Histogram

System.Collections.CreateAddAndRemove<String>.HashSet(Size: 512)


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 45.192717967244704 > 43.46384640172371.
IsChangePoint: Marked as a change because one of 5/28/2022 5:31:50 AM, 6/26/2022 6:11:06 PM, 7/5/2022 10:15:28 AM falls between 6/26/2022 6:11:06 PM and 7/5/2022 10:15:28 AM.
IsRegressionStdDev: Marked as regression because -8.712660750977818 (T) = (0 -43883.96099923706) / Math.Sqrt((1445959.5460368379 / (10)) + (1540844.7893621535 / (37))) is less than -2.0141033888794695 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (10) + (37) - 2, .025) and -0.09370957198324821 = (40123.96172016789 - 43883.96099923706) / 40123.96172016789 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

@mrsharm
Copy link
Member

mrsharm commented Aug 12, 2022

Hi @kunalspathak,

Seems like the regression persisted and was seemingly caused by the PGO update based on the historical data for x86:

image

Should we open a new issue or do we consider this regression to be by design?

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

Result Ratio Alloc Delta Operating System Bit Processor Name
Slower 0.89 +0 Windows 11 Arm64 Microsoft SQ1 3.0 GHz
Same 1.06 +0 Windows 11 Arm64 Microsoft SQ1 3.0 GHz
Same 1.13 +0 macOS Monterey 12.3 Arm64 Apple M1 Max
Same 0.91 +0 Windows 10 X64 Intel Xeon CPU E5-1650 v4 3.60GHz
Slower 0.86 +0 Windows 10 X64 Intel Core i7-6700 CPU 3.40GHz (Skylake)
Slower 0.86 +0 Windows 10 X64 Intel Core i7-6700 CPU 3.40GHz (Skylake)
Same 1.11 +0 Windows 10 X64 Intel Core i7-8650U CPU 1.90GHz (Kaby Lake R)
Same 1.12 +0 Windows 10 X64 Intel Core i9-10900K CPU 3.70GHz
Slower 0.89 +0 Windows 11 X64 AMD Ryzen Threadripper PRO 3945WX 12-Cores
Slower 0.88 +0 Windows 11 X64 AMD Ryzen 9 3950X
Slower 0.82 +0 Windows 11 X64 AMD Ryzen 9 5900X
Slower 0.80 +0 Windows 11 X64 AMD Ryzen 9 5950X
Slower 0.85 +0 Windows 11 X64 Intel Core i7-8700 CPU 3.20GHz (Coffee Lake)
Same 1.11 +0 Windows 11 X64 Intel Core i9-10900K CPU 3.70GHz
Slower 0.88 +0 Windows 11 X64 11th Gen Intel Core i9-11900H 2.50GHz
Slower 0.88 +0 ubuntu 18.04 X64 Intel Xeon CPU E5-1650 v4 3.60GHz
Same 1.03 +0 ubuntu 18.04 X64 Intel Core i7-2720QM CPU 2.20GHz (Sandy Bridge)
Slower 0.80 +0 ubuntu 18.04 X64 Intel Core i7-8700 CPU 3.20GHz (Coffee Lake)
Same 1.07 +0 ubuntu 20.04 X64 AMD Ryzen 9 5900X
Slower 0.86 +0 ubuntu 20.04 X64 Intel Core i9-10900K CPU 3.70GHz
Same 0.98 +0 Windows 10 X86 Intel Xeon CPU E5-1650 v4 3.60GHz
Same 0.94 +0 Windows 10 X86 Intel Core i7-6700 CPU 3.40GHz (Skylake)
Slower 0.87 +0 Windows 11 X86 AMD Ryzen Threadripper PRO 3945WX 12-Cores
Slower 0.87 +0 macOS Big Sur 11.6.8 X64 Intel Core i5-4278U CPU 2.60GHz (Haswell)
Slower 0.86 +0 macOS Monterey 12.3.1 X64 Intel Core i7-5557U CPU 3.10GHz (Broadwell)
Same 0.99 +0 macOS Monterey 12.4 X64 Intel Core i5-4278U CPU 2.60GHz (Haswell)

@kunalspathak
Copy link
Member

The 3 spikes are something that is worth investigating:
image

  1. dotnet/runtime@52f4e7c...8fe39db
  2. dotnet/runtime@7331be5...9768606
  3. dotnet/runtime@9f68db2...214ca6d

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
arch-x86 branch-refs/heads/main kind-micro os-windows perf-regression PGO Applied if there were any profile guided optimization updates in the observed interval. runtime-coreclr untriaged
Projects
None yet
Development

No branches or pull requests

2 participants