[Perf] Linux/x64: 2 Regressions on 1/31/2024 11:26:48 AM

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	ca46ae5e88965d284a5ebdda29c9e40604812d09
Compare	3bf9beecd5f68a59fc407488ce8120ee3d4240c7
Diff	Diff
Configs	CompilationMode:wasm, RunKind:micro

Regressions in System.Numerics.Tests.Perf_VectorConvert

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
Narrow_uint - Duration of single invocation 📝 - Benchmark Source 📈 - ADX Test Multi Config Graph	28.28 μs	32.11 μs	1.14	0.00	False
Narrow_int - Duration of single invocation 📝 - Benchmark Source 📈 - ADX Test Multi Config Graph	28.17 μs	32.09 μs	1.14	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.Numerics.Tests.Perf_VectorConvert*'

Payloads

Baseline
Compare

System.Numerics.Tests.Perf_VectorConvert.Narrow_uint

ETL Files

Histogram

JIT Disasms

System.Numerics.Tests.Perf_VectorConvert.Narrow_int

ETL Files

Histogram

JIT Disasms

Docs

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

[radekdoulik]

@BrzVlad this is dotnet/runtime#97514

[BrzVlad]

Tested both linux-amd64 and wasm and couldn't reproduce regression. Probably this is just noise.

[kg]

This is resolved now, fwiw