[Perf] Linux/x64: 35 Regressions on 10/11/2024 8:14:13 PM

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_Array

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ArrayResize - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	37.91 ns	64.42 ns	1.70	0.02	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_Array*'

Details

System.Tests.Perf_Array.ArrayResize

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.IO.Tests.StreamReaderReadToEndTests

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	50.53 ms	66.01 ms	1.31	0.00	True
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	50.99 ms	65.54 ms	1.29	0.00	True
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	47.67 ms	61.99 ms	1.30	0.01	True
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	47.22 ms	59.17 ms	1.25	0.01	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	46.06 ms	60.59 ms	1.32	0.01	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	49.97 ms	64.49 ms	1.29	0.00	True
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	50.30 ms	63.99 ms	1.27	0.01	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	46.68 ms	60.88 ms	1.30	0.01	True
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	47.31 ms	60.95 ms	1.29	0.02	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	47.74 ms	61.26 ms	1.28	0.00	True
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	47.21 ms	60.98 ms	1.29	0.01	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	47.12 ms	59.46 ms	1.26	0.00	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	49.34 ms	63.46 ms	1.29	0.00	True
ReadToEndAsync - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	49.36 ms	63.32 ms	1.28	0.02	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	50.31 ms	63.86 ms	1.27	0.00	True
ReadToEnd - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	46.69 ms	59.81 ms	1.28	0.01	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.IO.Tests.StreamReaderReadToEndTests*'

Details

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [ 1, 8])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [ 1, 1])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [ 33, 128])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [1025, 2048])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [1025, 2048])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [ 1, 1])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [ 9, 32])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [ 33, 128])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [ 0, 1024])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [ 0, 0])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [ 129, 1024])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [ 0, 1024])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [ 9, 32])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEndAsync(LineLengthRange: [ 0, 0])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [ 1, 8])

ETL Files

Histogram

JIT Disasms

System.IO.Tests.StreamReaderReadToEndTests.ReadToEnd(LineLengthRange: [ 129, 1024])

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in BenchmarksGame.KNucleotide_9

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
RunBench - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	41.92 ms	60.98 ms	1.45	0.25	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'BenchmarksGame.KNucleotide_9*'

Details

BenchmarksGame.KNucleotide_9.RunBench

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in ByteMark

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
BenchStringSort - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	970.09 ms	1.06 secs	1.09	0.00	False
BenchLUDecomp - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	655.62 ms	746.87 ms	1.14	0.08	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'ByteMark*'

Details

ByteMark.BenchStringSort

ETL Files

Histogram

JIT Disasms

ByteMark.BenchLUDecomp

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in LinqBenchmarks

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Where01LinqMethodX - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	106.76 ms	117.20 ms	1.10	0.16	False
Where00LinqMethodX - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	196.51 ms	228.85 ms	1.16	0.12	False
Where00LinqQueryX - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	184.58 ms	233.31 ms	1.26	0.10	False
Where01LinqQueryX - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	108.13 ms	118.42 ms	1.10	0.12	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'LinqBenchmarks*'

Details

LinqBenchmarks.Where01LinqMethodX

ETL Files

Histogram

JIT Disasms

LinqBenchmarks.Where00LinqMethodX

ETL Files

Histogram

JIT Disasms

LinqBenchmarks.Where00LinqQueryX

ETL Files

Histogram

JIT Disasms

LinqBenchmarks.Where01LinqQueryX

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_GC<Byte>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
AllocateUninitializedArray - Duration of single invocation ADX - Test Multi Config Graph	64.12 ns	90.84 ns	1.42	0.05	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_GC<Byte>*'

Details

System.Tests.Perf_GC<Byte>.AllocateUninitializedArray(length: 10000, pinned: False)

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Text.RegularExpressions.Tests.Perf_Regex_Common

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
CtorInvoke - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	158.95 μs	180.16 μs	1.13	0.14	False
CtorInvoke - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	143.00 μs	158.57 μs	1.11	0.08	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Text.RegularExpressions.Tests.Perf_Regex_Common*'

Details

System.Text.RegularExpressions.Tests.Perf_Regex_Common.CtorInvoke(Options: IgnoreCase, Compiled)

ETL Files

Histogram

JIT Disasms

System.Text.RegularExpressions.Tests.Perf_Regex_Common.CtorInvoke(Options: Compiled)

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in BenchmarksGame.BinaryTrees_2

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
RunBench - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	86.62 ms	93.75 ms	1.08	0.14	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'BenchmarksGame.BinaryTrees_2*'

Details

BenchmarksGame.BinaryTrees_2.RunBench

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Threading.Tests.Perf_Timer

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ScheduleManyThenDisposeMany - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	224.52 ms	281.25 ms	1.25	0.07	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Threading.Tests.Perf_Timer*'

Details

System.Threading.Tests.Perf_Timer.ScheduleManyThenDisposeMany

ETL Files

Histogram

JIT Disasms

Docs

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

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Collections.Concurrent.AddRemoveFromDifferentThreads<Int32>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ConcurrentStack - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	43.97 ms	73.13 ms	1.66	0.35	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.Concurrent.AddRemoveFromDifferentThreads<Int32>*'

Details

System.Collections.Concurrent.AddRemoveFromDifferentThreads<Int32>.ConcurrentStack(Size: 2000000)

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in BenchmarksGame.BinaryTrees_5

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
RunBench - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	56.40 ms	105.21 ms	1.87	0.48	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'BenchmarksGame.BinaryTrees_5*'

Details

BenchmarksGame.BinaryTrees_5.RunBench

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Text.RegularExpressions.Tests.Perf_Regex_Cache

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
IsMatch - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	49.25 ms	53.03 ms	1.08	0.01	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Text.RegularExpressions.Tests.Perf_Regex_Cache*'

Details

System.Text.RegularExpressions.Tests.Perf_Regex_Cache.IsMatch(total: 40000, unique: 1600, cacheSize: 800)

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in BenchmarksGame.BinaryTrees_6

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
RunBench - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	272.02 ms	513.31 ms	1.89	0.58	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'BenchmarksGame.BinaryTrees_6*'

Details

BenchmarksGame.BinaryTrees_6.RunBench

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_GC<Char>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
AllocateUninitializedArray - Duration of single invocation ADX - Test Multi Config Graph	72.54 ns	128.67 ns	1.77	0.06	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_GC<Char>*'

Details

System.Tests.Perf_GC<Char>.AllocateUninitializedArray(length: 10000, pinned: False)

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	ViperUbuntu
Baseline	cfc7c204c2317b5b4471c841ce4304fd74610822
Compare	a038f65e16afd2592c313bd2ee7dbcf4a00ae17b
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Collections.Concurrent.AddRemoveFromDifferentThreads<String>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ConcurrentStack - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	48.00 ms	76.02 ms	1.58	0.27	True

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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.Concurrent.AddRemoveFromDifferentThreads<String>*'

Details

System.Collections.Concurrent.AddRemoveFromDifferentThreads<String>.ConcurrentStack(Size: 2000000)

ETL Files

Histogram

JIT Disasms

Docs

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

[DrewScoggins]

Regressions likely caused by #108492.

[AndyAyersMS]

Also substantial numbers of improvements:

• [Perf] Linux/x64: 228 Improvements on 10/11/2024 8:14:13 PM perf-autofiling-issues#43221
• [Perf] Linux/x64: 16 Improvements on 10/11/2024 1:16:47 PM perf-autofiling-issues#43220
• [Perf] Linux/x64: 5 Improvements on 10/12/2024 12:44:56 AM perf-autofiling-issues#43222
• [Perf] Linux/x64: 5 Improvements on 10/10/2024 5:08:17 PM perf-autofiling-issues#43217

(not sure why these are different times)

[mrsharm]

Thanks for reporting this @DrewScoggins! These regressions are somewhat expected on machines running on certain AMD processors SKUs running either in a VM or a container. Would you happen to know what processor these tests are run on specifically which SKU?

The fix itself ensures that we are reading the right last level cache size to discern the initial gen0 budget that affects the GC behavior significantly. Without this fix we were reading in the host's (rather than the VM's) last level cache size. My guess is that the regressions were run on the machines that were susceptible to these issues and are now retrieving the right cache size and have caused some significant regressions and improvements because the behavior of the GC has now been changed (made correct).

The work-around that we have provided is to set the environment variable: DOTNET_GCCacheSizeFromSysConf to 1 to revert to the old behavior.

Please let me know if you have any further questions; if not, I can close this issue.