[Perf] Regressions from inliner policy change

[performanceautofiler]

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Memory.ReadOnlySpan

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
IndexOfString - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	39.46 ns	45.49 ns	1.15	0.00	False
IndexOfString - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	250.88 ns	300.80 ns	1.20	0.11	False
IndexOfString - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	22.82 ns	28.21 ns	1.24	0.00	False
IndexOfString - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	30.78 ns	36.42 ns	1.18	0.00	False
IndexOfString - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	27.12 ns	32.08 ns	1.18	0.00	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.Memory.ReadOnlySpan*'

System.Memory.ReadOnlySpan.IndexOfString(input: "foobardzsdzs", value: "rddzs", comparisonType: InvariantCulture)

ETL Files

Histogram

JIT Disasms

System.Memory.ReadOnlySpan.IndexOfString(input: "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAXAAAAAAAAAAAAAAAAAAAAAAAAAAAAA", value: "x", comparisonType: InvariantCultureIgnoreCase)

ETL Files

Histogram

JIT Disasms

System.Memory.ReadOnlySpan.IndexOfString(input: "ABCDE", value: "c", comparisonType: InvariantCultureIgnoreCase)

ETL Files

Histogram

JIT Disasms

System.Memory.ReadOnlySpan.IndexOfString(input: "string1", value: "string2", comparisonType: InvariantCulture)

ETL Files

Histogram

JIT Disasms

System.Memory.ReadOnlySpan.IndexOfString(input: "AAAAA5AAAA", value: "5", comparisonType: InvariantCulture)

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Text.Json.Tests.Perf_Basic

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
WriteBasicUtf16 - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.82 ms	2.02 ms	1.11	0.01	False
WriteBasicUtf8 - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	569.82 ns	632.04 ns	1.11	0.00	False
WriteBasicUtf8 - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.78 ms	2.04 ms	1.15	0.00	False
WriteBasicUtf8 - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	606.24 ns	661.19 ns	1.09	0.00	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.Json.Tests.Perf_Basic*'

System.Text.Json.Tests.Perf_Basic.WriteBasicUtf16(Formatted: False, SkipValidation: True, DataSize: 100000)

ETL Files

Histogram

JIT Disasms

System.Text.Json.Tests.Perf_Basic.WriteBasicUtf8(Formatted: False, SkipValidation: True, DataSize: 10)

ETL Files

Histogram

JIT Disasms

System.Text.Json.Tests.Perf_Basic.WriteBasicUtf8(Formatted: False, SkipValidation: True, DataSize: 100000)

ETL Files

Histogram

JIT Disasms

System.Text.Json.Tests.Perf_Basic.WriteBasicUtf8(Formatted: False, SkipValidation: False, DataSize: 10)

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_Guid

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ctor_str - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	33.19 ns	35.91 ns	1.08	0.01	False
ParseExactD - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	33.05 ns	36.90 ns	1.12	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.Tests.Perf_Guid*'

System.Tests.Perf_Guid.ctor_str

ETL Files

Histogram

JIT Disasms

System.Tests.Perf_Guid.ParseExactD

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Count - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	231.75 μs	244.13 μs	1.05	0.00	False
Count - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	158.99 μs	166.98 μs	1.05	0.00	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_Industry_RustLang_Sherlock*'

System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Holmes", Options: None)

ETL Files

Histogram

JIT Disasms

System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Sherlock Holmes", Options: NonBacktracking)

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Buffers.Tests.RentReturnArrayPoolTests<Byte>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
MultipleSerial - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	337.73 ns	407.52 ns	1.21	0.03	False
SingleParallel - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	54.42 μs	61.41 μs	1.13	0.43	False
SingleParallel - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	8.46 μs	9.47 μs	1.12	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Buffers.Tests.RentReturnArrayPoolTests<Byte>*'

System.Buffers.Tests.RentReturnArrayPoolTests<Byte>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: False)

ETL Files

Histogram

JIT Disasms

System.Buffers.Tests.RentReturnArrayPoolTests<Byte>.SingleParallel(RentalSize: 4096, ManipulateArray: True, Async: False, UseSharedPool: False)

ETL Files

Histogram

JIT Disasms

System.Buffers.Tests.RentReturnArrayPoolTests<Byte>.SingleParallel(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: False)

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Text.Json.Document.Tests.Perf_ParseThenWrite

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ParseThenWrite - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	19.70 μs	22.00 μs	1.12	0.04	False
ParseThenWrite - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	13.30 μs	16.23 μs	1.22	0.01	False
ParseThenWrite - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.39 ms	1.65 ms	1.19	0.28	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.Json.Document.Tests.Perf_ParseThenWrite*'

System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: True, TestCase: DeepTree)

ETL Files

Histogram

JIT Disasms

System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: False, TestCase: Json4KB)

ETL Files

Histogram

JIT Disasms

System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: False, TestCase: Json400KB)

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Globalization.Tests.StringSearch

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
IndexOf_Word_NotFound - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	821.17 ns	898.62 ns	1.09	0.01	False
IndexOf_Word_NotFound - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	541.36 ns	614.16 ns	1.13	0.11	False
IndexOf_Word_NotFound - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	539.89 ns	613.38 ns	1.14	0.11	False
IndexOf_Word_NotFound - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	821.32 ns	901.57 ns	1.10	0.01	False
IndexOf_Word_NotFound - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	540.29 ns	613.34 ns	1.14	0.10	False
LastIndexOf_Word_NotFound - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	852.78 ns	926.97 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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Globalization.Tests.StringSearch*'

System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, IgnoreCase, False))

ETL Files

Histogram

JIT Disasms

System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (, None, False))

ETL Files

Histogram

JIT Disasms

System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, IgnoreNonSpace, False))

ETL Files

Histogram

JIT Disasms

System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (, IgnoreCase, False))

ETL Files

Histogram

JIT Disasms

System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, None, False))

ETL Files

Histogram

JIT Disasms

System.Globalization.Tests.StringSearch.LastIndexOf_Word_NotFound(Options: (en-US, IgnoreCase, False))

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Buffers.Tests.RentReturnArrayPoolTests<Object>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
MultipleSerial - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	339.43 ns	398.33 ns	1.17	0.02	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.Buffers.Tests.RentReturnArrayPoolTests<Object>*'

System.Buffers.Tests.RentReturnArrayPoolTests<Object>.MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: False)

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Text.Json.Tests.Utf8JsonReaderCommentsTests

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Utf8JsonReaderCommentParsing - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	4.41 μs	4.78 μs	1.08	0.00	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.Json.Tests.Utf8JsonReaderCommentsTests*'

System.Text.Json.Tests.Utf8JsonReaderCommentsTests.Utf8JsonReaderCommentParsing(CommentHandling: Skip, SegmentSize: 0, TestCase: LongMultiLine)

ETL Files

Histogram

JIT Disasms

Docs

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

---

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Globalization.Tests.StringEquality

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Compare_Same - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.38 μs	1.72 μs	1.25	0.00	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.Globalization.Tests.StringEquality*'

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

ETL Files

Histogram

JIT Disasms

Docs

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

[performanceautofiler]

Run Information

Name	Value
Architecture	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Collections.Sort<BigStruct>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
LinqOrderByExtension - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	20.39 μs	22.32 μs	1.09	0.11	False
LinqQuery - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	20.54 μs	21.97 μs	1.07	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.git
python3 .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Collections.Sort<BigStruct>*'

System.Collections.Sort<BigStruct>.LinqOrderByExtension(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.Sort<BigStruct>.LinqQuery(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	arm64
OS	ubuntu 22.04
Queue	AmpereUbuntu
Baseline	27d23218a56d484638692d56b3abd27e0b21b591
Compare	d9c4c3e73dcf09435a3cc1cabb23584c9f24b504
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Text.Json.Serialization.Tests.WriteJson<ImmutableSortedDictionary<String, String>>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
SerializeToUtf8Bytes - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	9.13 μs	10.34 μs	1.13	0.29	False
SerializeToString - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	10.27 μs	12.48 μs	1.21	0.28	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.Json.Serialization.Tests.WriteJson<ImmutableSortedDictionary<String, String>>*'

System.Text.Json.Serialization.Tests.WriteJson<ImmutableSortedDictionary<String, String>>.SerializeToUtf8Bytes(Mode: SourceGen)

ETL Files

Histogram

JIT Disasms

System.Text.Json.Serialization.Tests.WriteJson<ImmutableSortedDictionary<String, String>>.SerializeToString(Mode: SourceGen)

ETL Files

Histogram

JIT Disasms

Docs

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

[AndyAyersMS]

#114821

[AndyAyersMS]

Related regressions:

• [Perf] Linux/x64: 20 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54110
• [Perf] Windows/x64: 17 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54088
• [Perf] Linux/x64: 24 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54061
• [Perf] Windows/x64: 17 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54082
• [Perf] Windows/x64: 5 Regressions on 4/19/2025 12:13:01 AM +00:00 perf-autofiling-issues#54081
• [Perf] Windows/x64: 2 Regressions on 4/21/2025 9:32:06 AM +00:00 perf-autofiling-issues#54089
• [Perf] Linux/x64: 1 Regression on 4/21/2025 9:32:06 AM +00:00 perf-autofiling-issues#54062
• [Perf] Linux/x64: 1 Regression on 4/21/2025 9:32:06 AM +00:00 perf-autofiling-issues#54111
• [Perf] Windows/x64: 1 Regression on 4/18/2025 6:27:32 PM +00:00 perf-autofiling-issues#54086

More reports (possibly dups):

• [Perf] Linux/x64: 25 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54630
• [Perf] Windows/x64: 27 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54609
• [Perf] Windows/x64: 30 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54600
• [Perf] Linux/x64: 22 Regressions on 4/21/2025 2:03:19 AM +00:00 perf-autofiling-issues#54569
• [Perf] Linux/x64: 5 Regressions on 4/21/2025 9:32:06 AM +00:00 perf-autofiling-issues#54570
• [Perf] Windows/x64: 1 Regression on 4/21/2025 9:32:06 AM +00:00 perf-autofiling-issues#54610
• [Perf] Linux/x64: 1 Regression on 4/20/2025 10:05:57 PM +00:00 perf-autofiling-issues#54629

[dotnet-policy-service]

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

[AndyAyersMS]

It also looks like there is a substantial increase in ASP.NET app startup time, likely (but not yet confirmed) from this change. @sebastienros FYI

If so, we may need to be more aggressive about marking "noinline" for methods that are unlikely to ever be good inline candidates even under PGO. Seems like we could study the distribution of rejected/accepted discretionary inlines and perhaps spot some good critiera (eg what characterizes methods that are always rejected).

[AndyAyersMS]

It also looks like there is a substantial increase in ASP.NET app startup time, likely (but not yet confirmed) from this change. @sebastienros FYI

The "disabled PGO" run shows the same impact, and this change should only affect behavior when PGO is enabled. So perhaps the root cause is something else?

Commit range seems to be 8c3a2e1...e1702c3

Likewise the "full PGO" shows no change (disables R2R), if this was JIT related we'd expect it to show a slowdown too.

Wonder if somehow R2R images were invalidated and we've just ended jitting more methods at startup?

[AndyAyersMS]

Collated data

Notes	Recent Score	Orig Score	arm64 ubuntu	x64 ubuntu	x64 Windows	Viper Ubuntu	Viper Windows	Benchmark
	2.69	2.59			2.69 2.59			System.Globalization.Tests.StringEquality.Compare_Same_Upper(Count: 1024, Options: (en-US, OrdinalIgnoreCase))
	2.58	2.71				2.58 2.71		System.Buffers.Text.Tests.Utf8FormatterTests.FormatterInt32(value: 4)
	1.94	1.94			1.94 1.94			System.Globalization.Tests.StringSearch.LastIndexOf_Word_NotFound(Options: (en-US, None, False))
	1.94	1.63			1.94 1.63			System.Globalization.Tests.StringSearch.LastIndexOf_Word_NotFound(Options: (, None, False))
	1.76	1.14			1.76 1.14			System.Collections.Sort(Int32).Array_Comparison(Size: 512)
	1.62	1.73		1.62 1.73				System.Globalization.Tests.StringSearch.LastIndexOf_Word_NotFound(Options: (, IgnoreCase, False))
	1.59	1.39			1.59 1.39			System.Text.RegularExpressions.Tests.Perf_Regex_Industry_Leipzig.Count(Pattern: "\p{Sm}", Options: NonBacktracking)
	1.47	1.39		1.47 1.50				System.Buffers.Text.Tests.Utf8ParserTests.TryParseUInt64(value: 18446744073709551615)
	1.47	1.31		1.46 1.33				System.IO.Tests.Perf_Path.GetFullPathNoRedundantSegments
	1.46	1.45	1.25 1.25		1.69 1.69			System.Globalization.Tests.StringEquality.Compare_Same(Count: 1024, Options: (en-US, OrdinalIgnoreCase))
	1.41	1.36					1.43 1.38	System.Collections.CtorFromCollection(Int32).Stack(Size: 512)
	1.41	1.42		1.41 1.44				System.Memory.Span(Byte).LastIndexOfValue(Size: 512)
	1.37	1.39				1.37 1.39		System.Buffers.Text.Tests.Utf8FormatterTests.FormatterInt32(value: -2147483648)
	1.35	1.37				1.35 1.37		System.Buffers.Tests.RentReturnArrayPoolTests(Byte).SingleSerial(RentalSize: 4096, ManipulateArray: True, Async: True, UseSharedPool: True)
	1.35	1.36					1.36 1.38	System.Collections.CtorFromCollection(Int32).Queue(Size: 512)
	1.34	1.48		1.66 1.66		1.09 1.25		System.Collections.ContainsTrue(String).Queue(Size: 512)
	1.33	1.33	1.09 1.09	1.63 1.63				System.Globalization.Tests.StringSearch.LastIndexOf_Word_NotFound(Options: (en-US, IgnoreCase, False))
	1.33	1.33				1.34 1.34		System.Buffers.Text.Tests.Utf8ParserTests.TryParseDecimal(value: 123456.789)
	1.32	1.34		1.50 1.35		1.17 1.36		System.Collections.ContainsFalse(String).Queue(Size: 512)
	1.30	1.25				1.31 1.25		System.Buffers.Text.Tests.Utf8FormatterTests.FormatterDecimal(value: 123456.789)
	1.30	1.29				1.27 1.27		System.Text.Json.Tests.Perf_Get.GetDecimal
	1.29	1.09				1.29 1.09		System.Net.Sockets.Tests.SocketSendReceivePerfTest.ReceiveAsyncThenSendAsync_Task
	1.27	1.29					1.28 1.30	System.Collections.CtorFromCollection(Int32).List(Size: 512)
	1.27	1.28			1.19 1.21		1.38 1.38	System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: True, TestCase: Json4KB)
	1.25	1.26			1.25 1.26			System.Net.Tests.Perf_WebUtility.Decode_DecodingRequired
	1.25	1.23		1.25 1.23				System.Buffers.Tests.RentReturnArrayPoolTests(Byte).SingleSerial(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: True)
	1.24	1.26			1.16 1.17		1.34 1.36	System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: True, TestCase: Json400KB)
	1.24	1.26				1.24 1.26		System.Buffers.Text.Tests.Utf8FormatterTests.FormatterInt32(value: 2147483647)
	1.24	1.22		1.24 1.23				System.IO.Tests.BinaryWriterExtendedTests.WriteAsciiString(StringLengthInChars: 8000)
	1.23	1.23		1.23 1.26				System.IO.Tests.BinaryWriterExtendedTests.WriteAsciiCharArray(StringLengthInChars: 8000)
	1.23	1.23		1.30 1.31	1.16 1.15			Benchmark.GetChildKeysTests.AddChainedConfigurationEmpty
	1.22	1.27		1.22 1.27				System.Buffers.Text.Tests.Utf8ParserTests.TryParseUInt64(value: 12345)
	1.22	1.22			1.22 1.22			System.Collections.Sort(Int32).LinqQuery(Size: 512)
	1.20	1.19					1.19 1.18	System.Collections.CtorFromCollection(Int32).ImmutableArray(Size: 512)
	1.19	1.19			1.19 1.18			System.Globalization.Tests.StringSearch.IsPrefix_FirstHalf(Options: (en-US, Ordinal, False))
	1.18	1.19		1.18 1.20				System.IO.Tests.Perf_Path.GetFullPathForLegacyLength
	1.17	1.13	1.17 1.13					System.Buffers.Tests.RentReturnArrayPoolTests(Byte).SingleParallel(RentalSize: 4096, ManipulateArray: True, Async: False, UseSharedPool: False)
	1.17	1.14			1.17 1.14			System.Memory.Span(Char).LastIndexOfValue(Size: 512)
	1.16	1.13	1.18 1.19	1.12 1.16	1.26 1.10	1.11 1.12		System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: False, TestCase: Json400KB)
	1.16	1.15					1.18 1.18	System.Text.Tests.Perf_Encoding.GetBytes(size: 512, encName: "utf-8")
	1.16	1.13	1.13 1.15	1.07 1.06	1.20 1.21	1.30 1.15	1.14 1.14	System.Text.Json.Tests.Perf_Basic.WriteBasicUtf8(Formatted: False, SkipValidation: True, DataSize: 100000)
	1.16	1.15		1.22 1.17		1.06 1.09		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_SliceSlice.Count(Options: IgnoreCase, NonBacktracking)
	1.16	1.11			1.16 1.11			System.Collections.Sort(Int32).LinqOrderByExtension(Size: 512)
	1.16	1.15				1.16 1.17		System.Tests.Perf_Decimal.ToString(value: 123456.789)
	1.15	1.17	1.22 1.22	1.17 1.23	1.12 1.15	1.18 1.20	1.16 1.17	System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: False, TestCase: Json4KB)
	1.15	1.12	1.06 1.12	1.25 1.12				System.Buffers.Tests.RentReturnArrayPoolTests(Byte).SingleParallel(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: False)
	1.15	1.37		1.13 1.45		1.13 1.31		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Sherlock", Options: NonBacktracking)
	1.15	1.10		1.15 1.10				System.Buffers.Tests.RentReturnArrayPoolTests(Object).SingleParallel(RentalSize: 4096, ManipulateArray: True, Async: True, UseSharedPool: False)
	1.14	1.11		1.14 1.11		1.13 1.11		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_Leipzig.Count(Pattern: "(?i)Twain", Options: None)
	1.14	1.25		1.18 1.12		1.09 1.25		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Sherlock", Options: None)
	1.14	1.14					1.13 1.13	System.IO.Tests.Perf_Directory.GetCurrentDirectory
	1.14	1.18					1.13 1.18	System.Buffers.Tests.RentReturnArrayPoolTests(Byte).SingleParallel(RentalSize: 4096, ManipulateArray: True, Async: True, UseSharedPool: True)
	1.13	1.14				1.14 1.17		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_SliceSlice.Count(Options: IgnoreCase)
	1.13	1.14	1.13 1.14					System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, IgnoreNonSpace, False))
	1.13	1.13	1.13 1.13					System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (, None, False))
	1.13	1.14	1.13 1.14					System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, None, False))
	1.13	1.14			1.12 1.14	1.11 1.11	1.15 1.16	System.Text.Json.Tests.Perf_Basic.WriteBasicUtf8(Formatted: False, SkipValidation: False, DataSize: 100000)
	1.13	1.14		1.13 1.19				System.Collections.CreateAddAndRemove(String).Queue(Size: 512)
	1.12	1.12	1.12 1.12	1.07 1.07	1.12 1.13	1.11 1.14	1.21 1.20	System.Text.Json.Document.Tests.Perf_ParseThenWrite.ParseThenWrite(IsDataIndented: True, TestCase: DeepTree)
	1.12	1.13					1.11 1.14	System.Collections.Tests.Perf_BitArray.BitArrayIntArrayCtor(Size: 512)
	1.12	1.12			1.12 1.12			System.Collections.Sort(IntClass).Array_ComparerClass(Size: 512)
	1.12	1.07					1.12 1.07	System.Tests.Perf_Type.GetType_FullyQualifiedNames(input: typeof(System.Collections.Generic.Dictionary`2[]))
	1.12	1.08		1.16 1.07	1.11 1.11		1.09 1.09	System.Tests.Perf_Double.ToStringWithFormat(value: 1.7976931348623157E+308, format: "F50")
	1.12	1.18					1.12 1.18	System.Text.Json.Serialization.Tests.WriteJson(ImmutableSortedDictionary(String, String)).SerializeToString(Mode: SourceGen)
	1.12	1.17	1.14 1.24			1.09 1.11		System.Memory.ReadOnlySpan.IndexOfString(input: "ABCDE", value: "c", comparisonType: InvariantCultureIgnoreCase)
	1.12	1.54		1.25 1.34				System.Buffers.Tests.RentReturnArrayPoolTests(Object).SingleParallel(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: False)
	1.11	1.11			1.11 1.11			System.Buffers.Text.Tests.Utf8ParserTests.TryParseUInt32Hex(value: FFFFFFFFFFFFFFFF)
	1.11	1.12	1.11 1.12					System.Tests.Perf_Guid.ParseExactD
	1.11	1.18	1.11 1.18					System.Memory.ReadOnlySpan.IndexOfString(input: "AAAAA5AAAA", value: "5", comparisonType: InvariantCulture)
	1.11	1.10			1.12 1.11			System.Collections.Sort(IntClass).Array_Comparison(Size: 512)
	1.10	1.10	1.11 1.11		1.11 1.10			System.Text.Json.Tests.Perf_Basic.WriteBasicUtf8(Formatted: False, SkipValidation: True, DataSize: 10)
	1.10	1.13	1.05 1.05	1.17 1.18		1.04 1.09		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Sherlock Holmes", Options: NonBacktracking)
	1.10	1.12					1.10 1.12	System.Text.Json.Tests.Perf_Basic.WriteBasicUtf16(Formatted: False, SkipValidation: False, DataSize: 100000)
	1.10	1.11	1.10 1.11		1.13 1.18	1.07 1.07	1.09 1.10	System.Text.Json.Tests.Perf_Basic.WriteBasicUtf16(Formatted: False, SkipValidation: True, DataSize: 100000)
	1.10	1.12	1.10 1.15		1.10 1.10			System.Memory.ReadOnlySpan.IndexOfString(input: "foobardzsdzs", value: "rddzs", comparisonType: InvariantCulture)
	1.10	1.20	1.10 1.20					System.Memory.ReadOnlySpan.IndexOfString(input: "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAXAAAAAAAAAAAAAAAAAAAAAAAAAAAAA", value: "x", comparisonType: InvariantCultureIgno
	1.10	1.08					1.10 1.08	System.Tests.Perf_Type.GetType_FullyQualifiedNames(input: typeof(System.Tests.NestedGeneric(String, Boolean)))
	1.10	1.12	1.12 1.18		1.06 1.07	1.08 1.08		System.Memory.ReadOnlySpan.IndexOfString(input: "string1", value: "string2", comparisonType: InvariantCulture)
	1.09	1.14		1.09 1.14				System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Holmes", Options: NonBacktracking)
	1.09	1.10	1.09 1.10					System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (, IgnoreCase, False))
	1.09	1.12		1.09 1.12				System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "zqj", Options: NonBacktracking)
	1.09	1.09	1.09 1.09					System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, IgnoreCase, False))
	1.09	1.12	1.15 1.17				1.04 1.07	System.Buffers.Tests.RentReturnArrayPoolTests(Object).MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: False)
	1.09	1.08	1.09 1.08					System.Text.Json.Tests.Utf8JsonReaderCommentsTests.Utf8JsonReaderCommentParsing(CommentHandling: Skip, SegmentSize: 0, TestCase: LongMultiLine)
	1.08	1.08					1.06 1.08	System.Reflection.Activator(EmptyClass).CreateInstanceNames
	1.08	1.12		1.08 1.10	1.10 1.28		1.08 1.08	System.Tests.Perf_Double.ToStringWithFormat(value: -1.7976931348623157E+308, format: "F50")
	1.08	1.09			1.08 1.09			System.Tests.Perf_Enum.InterpolateIntoSpan_Flags(value: Red, Green)
	1.08	1.08	1.08 1.08					System.Tests.Perf_Guid.ctor_str
	1.08	1.08	1.08 1.09		1.07 1.07			System.Text.Json.Tests.Perf_Basic.WriteBasicUtf8(Formatted: False, SkipValidation: False, DataSize: 10)
	1.07	1.08	1.05 1.05	1.10 1.12				System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Holmes", Options: None)
	1.07	1.06		1.09 1.06		1.06 1.06		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_Leipzig.Count(Pattern: "(?i)Twain", Options: NonBacktracking)
	1.07	1.07			1.09 1.06		1.06 1.09	System.Collections.Sort(IntClass).LinqOrderByExtension(Size: 512)
	1.07	1.06				1.07 1.06		System.Buffers.Tests.RentReturnArrayPoolTests(Byte).SingleParallel(RentalSize: 4096, ManipulateArray: False, Async: True, UseSharedPool: False)
	1.07	1.12	1.18 1.21	0.99 1.08		1.09 1.10	1.03 1.08	System.Buffers.Tests.RentReturnArrayPoolTests(Byte).MultipleSerial(RentalSize: 4096, ManipulateArray: False, Async: False, UseSharedPool: False)
	1.06	1.05			1.06 1.05			System.Perf_Convert.ChangeType
	1.06	1.07					1.06 1.07	System.Threading.Tests.Perf_CancellationToken.CreateLinkedTokenSource3
	1.06	1.07					1.06 1.07	System.Threading.Channels.Tests.BoundedChannelPerfTests.PingPong
	1.05	1.05			1.05 1.05			System.Perf_Convert.ToBase64String(formattingOptions: InsertLineBreaks)
	1.05	1.17		1.08 1.18		1.02 1.21		System.Text.RegularExpressions.Tests.Perf_Regex_Industry_RustLang_Sherlock.Count(Pattern: "(?i)Sherlock Holmes", Options: None)
	1.03	1.08			1.03 1.08			System.Text.Tests.Perf_Encoding.GetByteCount(size: 512, encName: "utf-8")
	1.02	1.12					1.03 1.10	System.Reflection.Attributes.IsDefinedMethodBaseMiss
	1.01	1.57					1.01 1.57	MicroBenchmarks.Serializers.Xml_ToStream(IndexViewModel).XmlSerializer_
	0.96	1.17			0.96 1.17			System.Globalization.Tests.StringSearch.IndexOf_Word_NotFound(Options: (en-US, Ordinal, False))