Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Perf] Linux/x64: 9 Improvements on 5/25/2023 4:37:39 AM #18355

Open
performanceautofiler bot opened this issue May 30, 2023 · 1 comment
Open

[Perf] Linux/x64: 9 Improvements on 5/25/2023 4:37:39 AM #18355

performanceautofiler bot opened this issue May 30, 2023 · 1 comment

Comments

@performanceautofiler
Copy link

performanceautofiler bot commented May 30, 2023

Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 2ec9a0ca9b76cd504b3e8d191b78a59a77c136b2
Compare ceb52b3dd1c7b9ea8d20ad46e7269391d8189614
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Single>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsStaticBenchmark - Duration of single invocation 58.62 ns 11.87 ns 0.20 0.35 False
GetHashCodeBenchmark - Duration of single invocation 550.32 ns 395.18 ns 0.72 0.15 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Single&gt;*'

Payloads

Baseline
Compare

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Single>.EqualsStaticBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 11.870512391029319 < 58.623534881684506.
IsChangePoint: Marked as a change because one of 3/8/2023 9:36:36 PM, 5/24/2023 11:17:37 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 104.23227500448957 (T) = (0 -13.546387470960227) / Math.Sqrt((49.30457025637768 / (299)) + (0.7524696152982344 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.7859070028419674 = (63.273379562998805 - 13.546387470960227) / 63.273379562998805 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

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


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 395.1801758689072 < 519.7883463396897.
IsChangePoint: Marked as a change because one of 3/18/2023 7:37:20 PM, 3/20/2023 8:31:51 AM, 5/23/2023 1:41:12 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 18.85997393933837 (T) = (0 -390.563609399939) / Math.Sqrt((33992.910801882084 / (299)) + (20.54604396178117 / (17))) is greater than 1.9675476980854363 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (17) - 2, .975) and 0.3410699482906228 = (592.7239293256549 - 390.563609399939) / 592.7239293256549 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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


Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 2ec9a0ca9b76cd504b3e8d191b78a59a77c136b2
Compare ceb52b3dd1c7b9ea8d20ad46e7269391d8189614
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in System.Numerics.Tests.Perf_Vector2

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsBenchmark - Duration of single invocation 173.07 ns 19.86 ns 0.11 0.35 False
GetHashCodeBenchmark - Duration of single invocation 110.89 ns 53.29 ns 0.48 0.22 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Vector2*'

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_Vector2.EqualsBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 19.859719674330304 < 164.29184726885578.
IsChangePoint: Marked as a change because one of 3/1/2023 3:09:21 PM, 3/2/2023 11:15:04 AM, 5/18/2023 7:10:13 PM, 5/24/2023 11:17:37 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 17.28582355941055 (T) = (0 -19.686035561966325) / Math.Sqrt((2040.327271788776 / (299)) + (0.5098932281225674 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.6970505410251135 = (64.98125340305768 - 19.686035561966325) / 64.98125340305768 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector2.GetHashCodeBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 53.29002299126405 < 106.44187990327492.
IsChangePoint: Marked as a change because one of 3/17/2023 8:17:18 AM, 5/23/2023 1:41:12 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 30.309124960169687 (T) = (0 -53.24584459174664) / Math.Sqrt((1611.2553813332459 / (299)) + (1.3572671290275438 / (17))) is greater than 1.9675476980854363 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (17) - 2, .975) and 0.5710276684442415 = (124.12419327521512 - 53.24584459174664) / 124.12419327521512 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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


Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 2ec9a0ca9b76cd504b3e8d191b78a59a77c136b2
Compare ceb52b3dd1c7b9ea8d20ad46e7269391d8189614
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in System.Numerics.Tests.Perf_Matrix4x4

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsBenchmark - Duration of single invocation 835.80 ns 86.58 ns 0.10 0.24 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Matrix4x4*'

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_Matrix4x4.EqualsBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 86.57920491226461 < 759.27593599005.
IsChangePoint: Marked as a change because one of 3/1/2023 3:09:21 PM, 3/2/2023 11:15:04 AM, 3/16/2023 12:15:41 AM, 3/16/2023 10:20:48 AM, 5/18/2023 7:10:13 PM, 5/24/2023 11:17:37 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 23.754656943860866 (T) = (0 -87.46933214653349) / Math.Sqrt((55848.21583784975 / (299)) + (17.533229302803136 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.7884088620618699 = (413.38844811217825 - 87.46933214653349) / 413.38844811217825 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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


Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 2ec9a0ca9b76cd504b3e8d191b78a59a77c136b2
Compare ceb52b3dd1c7b9ea8d20ad46e7269391d8189614
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in System.Numerics.Tests.Perf_Matrix3x2

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsBenchmark - Duration of single invocation 770.52 ns 81.54 ns 0.11 0.25 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Matrix3x2*'

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_Matrix3x2.EqualsBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 81.53712168020061 < 702.2081811420594.
IsChangePoint: Marked as a change because one of 3/1/2023 3:09:21 PM, 3/2/2023 11:15:04 AM, 5/18/2023 7:10:13 PM, 5/24/2023 11:17:37 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 15.872028838079665 (T) = (0 -81.12220816526865) / Math.Sqrt((29702.94220304633 / (299)) + (1.9922260786571877 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.6612154597349604 = (239.45073792861012 - 81.12220816526865) / 239.45073792861012 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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


Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 2ec9a0ca9b76cd504b3e8d191b78a59a77c136b2
Compare ceb52b3dd1c7b9ea8d20ad46e7269391d8189614
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in System.Numerics.Tests.Perf_Vector3

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsBenchmark - Duration of single invocation 175.82 ns 20.73 ns 0.12 0.31 False
GetHashCodeBenchmark - Duration of single invocation 149.52 ns 58.41 ns 0.39 0.22 False

graph
graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Vector3*'

Payloads

Baseline
Compare

Histogram

System.Numerics.Tests.Perf_Vector3.EqualsBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 20.73230811634801 < 167.66189152403246.
IsChangePoint: Marked as a change because one of 3/1/2023 3:09:21 PM, 3/2/2023 11:15:04 AM, 5/18/2023 7:10:13 PM, 5/24/2023 11:17:37 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 15.692557600515592 (T) = (0 -21.796845504085905) / Math.Sqrt((2104.0227643835096 / (299)) + (2.4041509325373607 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.6594938104047585 = (64.01306692837433 - 21.796845504085905) / 64.01306692837433 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

System.Numerics.Tests.Perf_Vector3.GetHashCodeBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 58.40914337961725 < 138.6953017318987.
IsChangePoint: Marked as a change because one of 3/15/2023 2:53:22 PM, 3/16/2023 1:49:47 PM, 5/23/2023 1:41:12 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 34.562146073493174 (T) = (0 -58.83486958545387) / Math.Sqrt((2720.527589417637 / (299)) + (3.773586060622856 / (17))) is greater than 1.9675476980854363 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (17) - 2, .975) and 0.6420205712897568 = (164.35265511604624 - 58.83486958545387) / 164.35265511604624 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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


Run Information

Name Value
Architecture x64
OS ubuntu 18.04
Queue TigerUbuntu
Baseline 2ec9a0ca9b76cd504b3e8d191b78a59a77c136b2
Compare ceb52b3dd1c7b9ea8d20ad46e7269391d8189614
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in System.Runtime.Intrinsics.Tests.Perf_Vector128Float

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio Baseline ETL Compare ETL
EqualsStaticBenchmark - Duration of single invocation 58.73 ns 14.13 ns 0.24 0.31 False

graph
Test Report

Repro

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

Payloads

Baseline
Compare

git clone https://github.com/dotnet/performance.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Float*'

Payloads

Baseline
Compare

Histogram

System.Runtime.Intrinsics.Tests.Perf_Vector128Float.EqualsStaticBenchmark


Description of detection logic

IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsRegressionBase: Marked as not a regression because the compare was not 5% greater than the baseline, or the value was too small.
IsImprovementBase: Marked as improvement because the compare was 5% less than the baseline, and the value was not too small.
IsImprovementCheck: Marked as improvement because the three check build points were 0.05 less than the baseline.
IsImprovementWindowed: Marked as improvement because 14.128791026358215 < 56.296652915921065.
IsChangePoint: Marked as a change because one of 3/9/2023 8:37:21 PM, 5/24/2023 11:17:37 PM, 5/29/2023 10:47:10 AM falls between 5/20/2023 9:58:12 PM and 5/29/2023 10:47:10 AM.
IsImprovementStdDev: Marked as improvement because 121.25005224168761 (T) = (0 -13.626696021055556) / Math.Sqrt((35.699113282536686 / (299)) + (0.5147789189518099 / (12))) is greater than 1.9676708853560316 = MathNet.Numerics.Distributions.StudentT.InvCDF(0, 1, (299) + (12) - 2, .975) and 0.7818788250002686 = (62.47305435188646 - 13.626696021055556) / 62.47305435188646 is greater than 0.05.
IsChangeEdgeDetector: Marked not as a regression because Edge Detector said so.

JIT Disasms

Docs

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

@lewing
Copy link
Member

lewing commented May 31, 2023

This is dotnet/runtime#86738

cc @kg

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

No branches or pull requests

1 participant