[Perf] Windows/x64: 15 Regressions on 5/8/2025 4:23:22 AM +00:00

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	Windows 10.0.22631
Queue	ViperWindows
Baseline	ac760bf5215a876647bea662df4ecc2e218dd221
Compare	177b1cf23a066e09719742ceae99dbc9744ee1dc
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Reflection.Attributes

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
IsDefinedMethodOverrideMiss - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	359.51 ns	385.12 ns	1.07	0.10	False
GetCustomAttributesMethodOverrideMiss - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	754.06 ns	822.34 ns	1.09	0.14	False
GetCustomAttributesMethodOverrideHitInherit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.21 μs	1.35 μs	1.11	0.07	False
GetCustomAttributesMethodBaseHitInherit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	765.31 ns	829.64 ns	1.08	0.11	False
IsDefinedMethodOverrideMissInherit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	361.64 ns	380.24 ns	1.05	0.12	False
GetCustomAttributesMethodOverrideHit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	747.49 ns	816.66 ns	1.09	0.10	False
IsDefinedMethodOverrideHitInherit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	348.85 ns	374.18 ns	1.07	0.14	False
IsDefinedClassHitInherit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	332.16 ns	359.43 ns	1.08	0.14	False
GetCustomAttributesClassHit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	708.78 ns	778.79 ns	1.10	0.12	False
IsDefinedMethodOverrideHit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	349.66 ns	389.89 ns	1.12	0.14	False
GetCustomAttributesMethodBaseHit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	755.04 ns	814.87 ns	1.08	0.10	False
GetCustomAttributesMethodOverrideMissInherit - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	748.58 ns	836.59 ns	1.12	0.15	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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Reflection.Attributes*'

System.Reflection.Attributes.IsDefinedMethodOverrideMiss

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.GetCustomAttributesMethodOverrideMiss

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.GetCustomAttributesMethodOverrideHitInherit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.GetCustomAttributesMethodBaseHitInherit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.IsDefinedMethodOverrideMissInherit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.GetCustomAttributesMethodOverrideHit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.IsDefinedMethodOverrideHitInherit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.IsDefinedClassHitInherit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.GetCustomAttributesClassHit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.IsDefinedMethodOverrideHit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.GetCustomAttributesMethodBaseHit

ETL Files

Histogram

JIT Disasms

System.Reflection.Attributes.GetCustomAttributesMethodOverrideMissInherit

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	Windows 10.0.22631
Queue	ViperWindows
Baseline	ac760bf5215a876647bea662df4ecc2e218dd221
Compare	177b1cf23a066e09719742ceae99dbc9744ee1dc
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in Interop.StructureToPtr

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
MarshalStructureToPtr - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	57.39 ns	70.54 ns	1.23	0.11	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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'Interop.StructureToPtr*'

Interop.StructureToPtr.MarshalStructureToPtr

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	Windows 10.0.22631
Queue	ViperWindows
Baseline	ac760bf5215a876647bea662df4ecc2e218dd221
Compare	177b1cf23a066e09719742ceae99dbc9744ee1dc
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Tests.Perf_Type

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
GetType_Resolvers - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	179.53 ns	205.80 ns	1.15	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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Tests.Perf_Type*'

System.Tests.Perf_Type.GetType_Resolvers(input: typeof(System.Int32[]))

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	Windows 10.0.22631
Queue	ViperWindows
Baseline	ac760bf5215a876647bea662df4ecc2e218dd221
Compare	177b1cf23a066e09719742ceae99dbc9744ee1dc
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Reflection.Invoke

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Field_SetStatic_struct - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	46.91 ns	54.61 ns	1.16	0.25	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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Reflection.Invoke*'

System.Reflection.Invoke.Field_SetStatic_struct

ETL Files

Histogram

JIT Disasms

Docs

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

[LoopedBard3]

Potentially just PGO related, so marking look again. Likely caused by: #115355 (1 of 2 PRs in range 1b5f930...b9d7c8b).
Related Regressions:

• [Perf] Linux/x64: 2 Regressions on 5/8/2025 4:23:22 AM +00:00 perf-autofiling-issues#55436
• [Perf] Windows/x64: 1 Regression on 5/8/2025 4:23:22 AM +00:00 perf-autofiling-issues#55424
• [Perf] Windows/arm64: 1 Regression on 5/8/2025 4:23:22 AM +00:00 perf-autofiling-issues#55572
• [Perf] Windows/arm64: 2 Regressions on 5/9/2025 2:58:35 AM +00:00 perf-autofiling-issues#55573
• [Perf] Windows/arm64: 4 Regressions on 5/8/2025 3:11:46 AM +00:00 perf-autofiling-issues#55940
• [Perf] Windows/arm64: 1 Regression on 5/6/2025 8:31:22 PM +00:00 perf-autofiling-issues#55939

[LoopedBard3]

Removing look again and transferring over as we are still seeing at least some of these regressions and have found a handful more.

[dotnet-policy-service]

Tagging subscribers to this area: @mangod9
See info in area-owners.md if you want to be subscribed.