Regression in multiple benchmarks from value numbering

[performanceautofiler]

Run Information

Architecture	x64
OS	Windows 10.0.19042
Baseline	9e5bb6526f968ceb2223fe81e2ad84fcc290c778
Compare	704f023147b49692cec842427b116e9ee7e64305
Diff	Diff

Regressions in System.Text.Encodings.Web.Tests.Perf_Encoders

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
EncodeUtf8 - Duration of single invocation	13.98 ns	15.58 ns	1.11	0.02	False

Test Report

Repro

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

Payloads

Baseline
Compare

Histogram

System.Text.Encodings.Web.Tests.Perf_Encoders.EncodeUtf8(arguments: UnsafeRelaxed,no <escaping /> required,16)

Docs

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

### Run Information

Architecture	x64
OS	Windows 10.0.19042
Baseline	9e5bb6526f968ceb2223fe81e2ad84fcc290c778
Compare	704f023147b49692cec842427b116e9ee7e64305
Diff	Diff

Regressions in LinqBenchmarks

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
Where00LinqQueryX - Duration of single invocation	626.59 ms	673.96 ms	1.08	0.05	False

Test Report

Repro

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

Payloads

Baseline
Compare

Histogram

LinqBenchmarks.Where00LinqQueryX

Docs

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

Run Information

Architecture	x64
OS	Windows 10.0.19042
Baseline	9e5bb6526f968ceb2223fe81e2ad84fcc290c778
Compare	704f023147b49692cec842427b116e9ee7e64305
Diff	Diff

Regressions in System.Collections.Tests.Perf_PriorityQueue<Int32, Int32>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
HeapSort - Duration of single invocation	2.69 μs	3.21 μs	1.19	0.07	False

Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Collections.Tests.Perf_PriorityQueue<Int32, Int32>*'

Payloads

Baseline
Compare

Histogram

System.Collections.Tests.Perf_PriorityQueue<Int32, Int32>.HeapSort(Size: 100)

Docs

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

Run Information

Architecture	x64
OS	Windows 10.0.19042
Baseline	9e5bb6526f968ceb2223fe81e2ad84fcc290c778
Compare	704f023147b49692cec842427b116e9ee7e64305
Diff	Diff

Regressions in System.IO.Tests.BinaryWriterExtendedTests

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
WriteAsciiCharArray - Duration of single invocation	79.79 μs	86.74 μs	1.09	0.05	False

Test Report

Repro

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

Payloads

Baseline
Compare

Histogram

System.IO.Tests.BinaryWriterExtendedTests.WriteAsciiCharArray(StringLengthInChars: 2000000)

Docs

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

Run Information

Architecture	x64
OS	Windows 10.0.19042
Baseline	9e5bb6526f968ceb2223fe81e2ad84fcc290c778
Compare	704f023147b49692cec842427b116e9ee7e64305
Diff	Diff

Regressions in System.Text.Json.Tests.Perf_Reader

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
ReadSpanEmptyLoop - Duration of single invocation	111.66 ns	126.26 ns	1.13	0.10	False

Test Report

Repro

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

Payloads

Baseline
Compare

Histogram

System.Text.Json.Tests.Perf_Reader.ReadSpanEmptyLoop(IsDataCompact: False, TestCase: HelloWorld)

Docs

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

Run Information

Architecture	x64
OS	Windows 10.0.19042
Baseline	d216c7c15cb0080342f1191368afc7e5be172de0
Compare	704f023147b49692cec842427b116e9ee7e64305
Diff	Diff

Regressions in System.Collections.ContainsTrue<String>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
Span - Duration of single invocation	358.31 μs	404.10 μs	1.13	0.01	False

Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Collections.ContainsTrue<String>*'

Payloads

Baseline
Compare

Histogram

System.Collections.ContainsTrue<String>.Span(Size: 512)

Docs

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

Run Information

Architecture	x64
OS	Windows 10.0.19042
Baseline	9e5bb6526f968ceb2223fe81e2ad84fcc290c778
Compare	704f023147b49692cec842427b116e9ee7e64305
Diff	Diff

Regressions in System.Buffers.Tests.ReadOnlySequenceTests<Char>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio	Baseline ETL	Compare ETL
FirstArray - Duration of single invocation	6.67 ns	9.94 ns	1.49	0.17	False

Test Report

Repro

git clone https://github.com/dotnet/performance.git
py .\performance\scripts\benchmarks_ci.py -f net6.0 --filter 'System.Buffers.Tests.ReadOnlySequenceTests<Char>*'

Payloads

Baseline
Compare

Histogram

System.Buffers.Tests.ReadOnlySequenceTests<Char>.FirstArray

Docs

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

[dotnet-issue-labeler]

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

[kunalspathak]

Possibly from 39ece73. @SingleAccretion

[SingleAccretion]

@kunalspathak did you mean to close this issue?

FWIW, I would not have expected 39ece73 to affect anything, seeing as there were almost no diffs.

[kunalspathak]

Sorry, I didn't mean to close the issue. The change point detection points to 39ece73. It will be good to double check and see if the result matches with and without the change.

[SingleAccretion]

I have checked all the relevant regressions against 39ece73 and was not able to reproduce them, or observe generated code differences.

[JulieLeeMSFT]

Thanks @SingleAccretion for checking your side of code.
@jakobbotsch for VN regression.

[jakobbotsch]

I doubt this is related to value numbering after what @SingleAccretion wrote, seems like the other suspect in the diff might be #61234 which is a VM change. I'll check.

[jakobbotsch]

From looking at the graphs it seems like all these cases are either back to normal now or are in a mode that we have seen before the commit hashes in the OP. Will close this.