[Perf] Linux/x64: 1 Regression on 9/10/2024 12:54:19 PM

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	76dbb27608765ac75fe2a06166271520fdab85fe
Compare	636c86342a29dfa8f419414bbd4877b882198ed3
Diff	Diff
Configs	CompilationMode:tiered, RunKind:micro

Regressions in System.Collections.TryGetValueFalse<Int32, Int32>

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
IDictionary - Duration of single invocation 📝 - Benchmark Source ADX - Test Multi Config Graph	2.00 μs	2.25 μs	1.13	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.TryGetValueFalse<Int32, Int32>*'

System.Collections.TryGetValueFalse<Int32, Int32>.IDictionary(Size: 512)

ETL Files

Histogram

JIT Disasms

Docs

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

[LoopedBard3]

Seeing this only on the two x64 Linux Queues, but no clear cause in the diffs: 4ee9789...ad6c18f.

[jeffschwMSFT]

@LoopedBard3 this test looks a bit noisy. have we seen this issue persist? I agree there is no obvious change that is the cause

[LoopedBard3]

The first link in the result table: "IDictionary - Duration of single invocation", links to the history for the test and is also kept up to date. There, it does appear that the bump up has persisted. At least as of 09/19.

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/area-system-runtime
See info in area-owners.md if you want to be subscribed.