[Perf] Linux/x64: 10 Regressions on 11/14/2023 8:48:07 AM

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	76aeefb8230c897ff83eaaa7fd66adbe9c7e6b5f
Compare	1e4d36da40bd0ad1ff36a398ccec6eea24540367
Diff	Diff
Configs	CompilationMode:wasm, RunKind:blazor_scenarios

Regressions in SOD - Minimum Blazor Template - Publish - AOT

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
SOD - Minimum Blazor Template - Publish - AOT - pub/wwwroot/_framework/System.Runtime.InteropServices.JavaScript.wasm.br 📈 - ADX Test Multi Config Graph	15.03 KB	14.83 KB	0.99	0.00	True
SOD - Minimum Blazor Template - Publish - AOT - pub - Count 📈 - ADX Test Multi Config Graph	124.00 count	155.00 count	1.25	0.00	True
SOD - Minimum Blazor Template - Publish - AOT - pub 📈 - ADX Test Multi Config Graph	21.46 MB	23.97 MB	1.12	0.00	True
SOD - Minimum Blazor Template - Publish - AOT - SOD - Minimum Blazor Template - Publish - AOT - Count 📈 - ADX Test Multi Config Graph	124.00 count	155.00 count	1.25	0.00	True
SOD - Minimum Blazor Template - Publish - AOT - SOD - Minimum Blazor Template - Publish - AOT 📈 - ADX Test Multi Config Graph	21.46 MB	23.97 MB	1.12	0.00	True

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 'SOD - Minimum Blazor Template - Publish - AOT*'

Payloads

Baseline
Compare

SOD - Minimum Blazor Template - Publish - AOT

ETL Files

Histogram

JIT Disasms

Docs

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

[performanceautofiler]

Run Information

Name	Value
Architecture	x64
OS	ubuntu 22.04
Queue	TigerUbuntu
Baseline	76aeefb8230c897ff83eaaa7fd66adbe9c7e6b5f
Compare	1e4d36da40bd0ad1ff36a398ccec6eea24540367
Diff	Diff
Configs	CompilationMode:wasm, RunKind:blazor_scenarios

Regressions in SOD - New Blazor Template - Publish - AOT

Benchmark	Baseline	Test	Test/Base	Test Quality	Edge Detector	Baseline IR	Compare IR	IR Ratio
SOD - New Blazor Template - Publish - AOT - pub/wwwroot/_framework/System.Runtime.InteropServices.JavaScript.wasm.br 📈 - ADX Test Multi Config Graph	15.03 KB	14.83 KB	0.99	0.00	True
SOD - New Blazor Template - Publish - AOT - pub 📈 - ADX Test Multi Config Graph	28.31 MB	30.85 MB	1.09	0.00	True
SOD - New Blazor Template - Publish - AOT - SOD - New Blazor Template - Publish - AOT - Count 📈 - ADX Test Multi Config Graph	133.00 count	164.00 count	1.23	0.00	True
SOD - New Blazor Template - Publish - AOT - SOD - New Blazor Template - Publish - AOT 📈 - ADX Test Multi Config Graph	28.31 MB	30.85 MB	1.09	0.00	True
SOD - New Blazor Template - Publish - AOT - pub - Count 📈 - ADX Test Multi Config Graph	133.00 count	164.00 count	1.23	0.00	True

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 'SOD - New Blazor Template - Publish - AOT*'

Payloads

Baseline
Compare

SOD - New Blazor Template - Publish - AOT

ETL Files

Histogram

JIT Disasms

Docs

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

[radekdoulik]

The range is dotnet/runtime@4287185...4f3bae5

Could this be caused by dotnet/runtime#90436?

@fanyang-mono @maraf how does it affect Blazor? Is it also enabled by default for Blazor? If not, could it cause size increase?

[fanyang-mono]

@radekdoulik What does SOD - Minimum Blazor Template - Publish - AOT - pub - Count measure?

[fanyang-mono]

dotnet/runtime#90436 didn't enable WasmStripILAfterAOT by default. I wouldn't expect any size changes after that specific PR. The PR to turn WasmStripILAfterAOT on by default is dotnet/runtime#94682.

That being said, dotnet/runtime#90436 contains changes regarding to Blazor app packaging. @maraf Could you investigate this first to see if any or your changes in the PR might caused this? If not, I could investigate further.