[Perf] System.IO.Compression Regressions on 10/21/2025 12:14:09 AM +00:00

[performanceautofiler]

Run Information

Name	Value
Architecture	arm64
OS	Windows 10.0.22621
Queue	AmpereWindows
Baseline	0d61284e49da7f38947a2af4363d1efc920bfc75
Compare	402d73579fd9a08972dd86f6fba034f23bf545e3
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.IO.Compression.Brotli

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Compress - Duration of single invocation ADX - Test Multi Config Graph	20.12 μs	1.23 ms	61.36	0.67	False
Compress - Duration of single invocation ADX - Test Multi Config Graph	550.67 μs	2.05 ms	3.71	0.46	False
Compress - Duration of single invocation ADX - Test Multi Config Graph	3.85 ms	4.28 ms	1.11	0.63	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.IO.Compression.Brotli*'

System.IO.Compression.Brotli.Compress(level: Optimal, file: "sum")

ETL Files

Histogram

JIT Disasms

System.IO.Compression.Brotli.Compress(level: Optimal, file: "TestDocument.pdf")

ETL Files

Histogram

JIT Disasms

System.IO.Compression.Brotli.Compress(level: Optimal, file: "alice29.txt")

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	Windows 10.0.22621
Queue	AmpereWindows
Baseline	0d61284e49da7f38947a2af4363d1efc920bfc75
Compare	402d73579fd9a08972dd86f6fba034f23bf545e3
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.IO.Compression.Gzip

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Compress - Duration of single invocation ADX - Test Multi Config Graph	3.17 ms	4.86 ms	1.53	0.59	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.IO.Compression.Gzip*'

System.IO.Compression.Gzip.Compress(level: Optimal, file: "TestDocument.pdf")

ETL Files

Histogram

JIT Disasms

Docs

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

[LoopedBard3]

From range: e7c8c1b...768bb6d, no clear cause. We are seeing this across all setups, including the seeming noise spike just before this regression. @AndyAyersMS @EgorBo is there a commit that sticks out to y'all? Potentially it is due to #119758?

Related Regressions:

• [Perf] Linux/arm64: 3 Regressions on 10/21/2025 12:14:09 AM +00:00 perf-autofiling-issues#64198
• [Perf] Windows/arm64: 1 Regression on 10/20/2025 2:32:30 PM +00:00 perf-autofiling-issues#64191 (seems to be noisy so datetime was misattributed)