dotnet/perf-autofiling-issuesPublic

NotificationsYou must be signed in to change notification settings
Fork4
Star10

[Perf] Linux/x64: 30 Regressions on 3/23/2024 2:15:48 AM #31868

Open

[Perf] Linux/x64: 30 Regressions on 3/23/2024 2:15:48 AM#31868

Labels

PGOApplied if there were any profile guided optimization updates in the observed interval.arch-wasmarch-x64branch-refs/heads/mainkind-microos-linuxperf-regressionuntriagedwasm-aot

Description

performanceautofiler

bot

opened

on Mar 26, 2024

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector
InequalityOperatorBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.00 ns	1.95 ns	∞	0.91	False
GreaterThanBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	3.79 ns	5.53 ns	1.46	0.83	False
LessThanAnyBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.63 ns	3.06 ns	4.82	0.92	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt32&gt;*'

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>.InequalityOperatorBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>.GreaterThanBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>.LessThanAnyBenchmark

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Collections.CtorGivenSize<Int32>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector
Array - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	188.20 ns	205.74 ns	1.09	0.06	False
SortedList - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	416.72 ns	458.70 ns	1.10	0.04	False
List - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	203.79 ns	227.25 ns	1.12	0.07	False
Stack - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	214.69 ns	235.03 ns	1.09	0.06	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.CtorGivenSize&lt;Int32&gt;*'

System.Collections.CtorGivenSize<Int32>.Array(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSize<Int32>.SortedList(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSize<Int32>.List(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSize<Int32>.Stack(Size: 512)

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector
DotBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.75 ns	1.83 ns	2.45	0.92	False
InequalityOperatorBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.00 ns	2.25 ns	∞	0.88	False
GreaterThanAllBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.31 ns	2.08 ns	6.70	0.90	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int32&gt;*'

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>.DotBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>.InequalityOperatorBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>.GreaterThanAllBenchmark

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

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

Benchmark	Test	Test/Base	Test Quality	Edge Detector
GreaterThanAnyBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	1.61 ns	∞	0.89	False
LessThanAnyBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	1.33 ns	∞	0.86	False
EqualsAnyBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	1.63 ns	∞	0.92	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Single&gt;*'

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

ETL Files

Histogram

JIT Disasms

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

ETL Files

Histogram

JIT Disasms

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

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Tests.Perf_GC<Byte>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
NewOperator_Array - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	76.70 ns	100.17 ns	1.31	0.10	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_GC&lt;Byte&gt;*'

System.Tests.Perf_GC<Byte>.NewOperator_Array(length: 1000)

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Text.Json.Tests.Perf_Get

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
GetByte - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	2.47 μs	2.91 μs	1.18	0.46	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Text.Json.Tests.Perf_Get*'

System.Text.Json.Tests.Perf_Get.GetByte

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector
EqualityOperatorBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.00 ns	2.15 ns	∞	0.92	False
GreaterThanAnyBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.16 ns	2.16 ns	13.63	0.93	False
InequalityOperatorBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.00 ns	1.83 ns	∞	0.86	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;UInt16&gt;*'

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>.EqualityOperatorBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>.GreaterThanAnyBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt16>.InequalityOperatorBenchmark

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Collections.CtorGivenSizeNonGeneric

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector
Stack - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	221.80 ns	239.18 ns	1.08	0.09	False
Queue - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	235.28 ns	250.13 ns	1.06	0.10	False
SortedList - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	476.91 ns	508.75 ns	1.07	0.03	False
ArrayList - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	210.21 ns	229.59 ns	1.09	0.09	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.CtorGivenSizeNonGeneric*'

System.Collections.CtorGivenSizeNonGeneric.Stack(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSizeNonGeneric.Queue(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSizeNonGeneric.SortedList(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSizeNonGeneric.ArrayList(Size: 512)

ETL Files

Histogram

JIT Disasms

Docs

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

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	a2ce8db5fb14df85cef03880a30133be198626e1
Compare	a264a7e4134a5a70bb0cc422bac5c232ba8ff97c
Diff	Diff
Configs	AOT:true, CompilationMode:wasm, RunKind:micro

Regressions in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int16>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
AllBitsSetBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	0.49 ns	4.09 ns	8.31	0.87	False
GreaterThanOrEqualBenchmark - Duration of single invocation 📝 -Benchmark Source ADX - Test Multi Config Graph	3.22 ns	5.96 ns	1.85	0.86	False

Test Report

Repro

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

git clone https://github.com/dotnet/performance.gitpython3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int16&gt;*'

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int16>.AllBitsSetBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int16>.GreaterThanOrEqualBenchmark

ETL Files

Histogram

JIT Disasms

Docs

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

Metadata

Assignees

No one assigned

Labels

PGOApplied if there were any profile guided optimization updates in the observed interval.arch-wasmarch-x64branch-refs/heads/mainkind-microos-linuxperf-regressionuntriagedwasm-aot

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Movatterモバイル変換

[Perf] Linux/x64: 30 Regressions on 3/23/2024 2:15:48 AM #31868

Description

Run Information

Regressions in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>

Repro

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>.InequalityOperatorBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>.GreaterThanBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<UInt32>.LessThanAnyBenchmark

ETL Files

Histogram

JIT Disasms

Docs

Run Information

Regressions in System.Collections.CtorGivenSize<Int32>

Repro

System.Collections.CtorGivenSize<Int32>.Array(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSize<Int32>.SortedList(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSize<Int32>.List(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.CtorGivenSize<Int32>.Stack(Size: 512)

ETL Files

Histogram

JIT Disasms

Docs

Run Information

Regressions in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>

Repro

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>.DotBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>.InequalityOperatorBenchmark

ETL Files

Histogram

JIT Disasms

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int32>.GreaterThanAllBenchmark

ETL Files

Histogram

JIT Disasms

Docs

Run Information

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

Repro

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

ETL Files

Histogram

JIT Disasms

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

ETL Files

Histogram

JIT Disasms

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

ETL Files

Histogram

JIT Disasms

Docs

Run Information

Regressions in System.Tests.Perf_GC<Byte>

Repro

System.Tests.Perf_GC<Byte>.NewOperator_Array(length: 1000)

ETL Files

Histogram

JIT Disasms

Docs

Run Information