[Perf] Perf Regressions from caching profile data for inlinee casts

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	6ab0311319cc7c975ac7ed1b84df809e881dfb13
Compare	f34ec2bc4c3a16b647fd4fc0016aca4f62479e5a
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in PerfLabTests.CastingPerf

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ObjFooIsObj - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	216.84 μs	240.74 μs	1.11	0.02	False
FooObjIsFoo2 - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	217.63 μs	240.64 μs	1.11	0.01	False
IFooObjIsIFooInterAlia - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	217.14 μs	241.31 μs	1.11	0.01	False
ObjObjrefValueType - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	216.96 μs	242.93 μs	1.12	0.02	False
ObjScalarValueType - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	218.05 μs	241.54 μs	1.11	0.03	False
FooObjIsNull - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	218.58 μs	240.94 μs	1.10	0.05	False
IFooFooIsIFoo - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	223.21 μs	243.98 μs	1.09	0.03	False
IFooObjIsIFoo - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	216.26 μs	240.73 μs	1.11	0.01	False
ObjInt - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	217.04 μs	241.42 μs	1.11	0.02	False
ObjFooIsObj2 - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	217.99 μs	240.90 μs	1.11	0.02	False
IFooObjIsDescendantOfIFoo - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	218.15 μs	240.80 μs	1.10	0.02	False
FooObjIsDescendant - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	216.98 μs	241.12 μs	1.11	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 'PerfLabTests.CastingPerf*'

PerfLabTests.CastingPerf.ObjFooIsObj

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.FooObjIsFoo2

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.IFooObjIsIFooInterAlia

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.ObjObjrefValueType

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.ObjScalarValueType

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.FooObjIsNull

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.IFooFooIsIFoo

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.IFooObjIsIFoo

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.ObjInt

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.ObjFooIsObj2

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.IFooObjIsDescendantOfIFoo

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf.FooObjIsDescendant

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	6ab0311319cc7c975ac7ed1b84df809e881dfb13
Compare	f34ec2bc4c3a16b647fd4fc0016aca4f62479e5a
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Collections.IterateForEach<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	2.10 μs	2.59 μs	1.23	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.Collections.IterateForEach<String>*'

System.Collections.IterateForEach<String>.ConcurrentStack(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	fcd197c9a4ea9def2049fdb302eede3e044fba5f
Compare	f34ec2bc4c3a16b647fd4fc0016aca4f62479e5a
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
Where00LinqMethodX - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	271.68 ms	287.71 ms	1.06	0.04	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*'

LinqBenchmarks.Where00LinqMethodX

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	6ab0311319cc7c975ac7ed1b84df809e881dfb13
Compare	f34ec2bc4c3a16b647fd4fc0016aca4f62479e5a
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in PerfLabTests.CastingPerf2.CastingPerf

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
IFooObjIsIFoo - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	219.63 μs	244.72 μs	1.11	0.02	False
FooObjIsDescendant - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	211.49 μs	241.66 μs	1.14	0.00	False
FooObjCastIfIsa - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	225.93 μs	329.76 μs	1.46	0.33	False
IFooFooIsIFoo - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	436.06 μs	484.38 μs	1.11	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 'PerfLabTests.CastingPerf2.CastingPerf*'

PerfLabTests.CastingPerf2.CastingPerf.IFooObjIsIFoo

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf2.CastingPerf.FooObjIsDescendant

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf2.CastingPerf.FooObjCastIfIsa

ETL Files

Histogram

JIT Disasms

PerfLabTests.CastingPerf2.CastingPerf.IFooFooIsIFoo

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	6ab0311319cc7c975ac7ed1b84df809e881dfb13
Compare	f34ec2bc4c3a16b647fd4fc0016aca4f62479e5a
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Perf_Convert

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
ToBase64String - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.17 μs	1.49 μs	1.27	0.31	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.Perf_Convert*'

System.Perf_Convert.ToBase64String(formattingOptions: InsertLineBreaks)

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	6ab0311319cc7c975ac7ed1b84df809e881dfb13
Compare	f34ec2bc4c3a16b647fd4fc0016aca4f62479e5a
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Xml.Linq.Perf_XElementList

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Enumerator - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	171.40 ns	184.22 ns	1.07	0.03	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.Xml.Linq.Perf_XElementList*'

System.Xml.Linq.Perf_XElementList.Enumerator

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	6ab0311319cc7c975ac7ed1b84df809e881dfb13
Compare	f34ec2bc4c3a16b647fd4fc0016aca4f62479e5a
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Collections.IterateForEachNonGeneric<String>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Queue - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.80 μs	1.99 μs	1.11	0.02	False
Stack - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	1.50 μs	1.62 μs	1.08	0.04	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.IterateForEachNonGeneric<String>*'

System.Collections.IterateForEachNonGeneric<String>.Queue(Size: 512)

ETL Files

Histogram

JIT Disasms

System.Collections.IterateForEachNonGeneric<String>.Stack(Size: 512)

ETL Files

Histogram

JIT Disasms

Docs

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

[LoopedBard3]

Likely #116241 from range: 6ab0311...ab30ec7. FYI @AndyAyersMS. It seems the greatest impact was by far on Tiger Linux.

Related Regressions:

• [Perf] Windows/x64: 4 Regressions on 6/3/2025 10:54:35 PM +00:00 perf-autofiling-issues#57165
• [Perf] Windows/x64: 7 Regressions on 6/3/2025 10:54:35 PM +00:00 perf-autofiling-issues#57115
• [Perf] Linux/arm64: 3 Regressions on 6/3/2025 8:09:30 PM +00:00 perf-autofiling-issues#57430

Improvements:

• [Perf] Linux/arm64: 15 Improvements on 6/3/2025 8:09:30 PM +00:00 perf-autofiling-issues#56958
• [Perf] Linux/x64: 113 Improvements on 6/3/2025 10:54:35 PM +00:00 perf-autofiling-issues#57258
• [Perf] Windows/x64: 93 Improvements on 6/3/2025 10:54:35 PM +00:00 perf-autofiling-issues#57217
• [Perf] Linux/x64: 103 Improvements on 6/3/2025 10:54:35 PM +00:00 perf-autofiling-issues#57143
• [Perf] Windows/x64: 36 Improvements on 6/3/2025 10:54:35 PM +00:00 perf-autofiling-issues#57133
• [Perf] Windows/arm64: 11 Improvements on 6/3/2025 11:51:36 PM +00:00 perf-autofiling-issues#57545
• [Perf] Windows/arm64: 6 Improvements on 6/3/2025 8:09:30 PM +00:00 perf-autofiling-issues#57544
• [Perf] Linux/arm64: 28 Improvements on 6/3/2025 8:09:30 PM +00:00 perf-autofiling-issues#57437
• [Perf] Linux/arm64: 14 Improvements on 6/3/2025 2:14:15 PM +00:00 perf-autofiling-issues#57436
• [Perf] Windows/arm64: 6 Improvements on 6/4/2025 4:21:45 AM +00:00 perf-autofiling-issues#57546

[AndyAyersMS]

Likely JCC related.