Repository navigation
[Perf] Windows/x64: Regressions in System.Collections on 8/2/2024 1:16:56 PM #106027
Description
Run Information
| Name | Value |
|---|---|
| Architecture | x64 |
| OS | Windows 10.0.22621 |
| Queue | TigerWindows |
| Baseline | 30b34e6d7b475c1de91623d3589723181a260523 |
| Compare | 3e572e9d09026f386a57302bfd624abd7b660fb5 |
| Diff | Diff |
| Configs | CompilationMode:tiered, RunKind:micro |
Regressions in System.Collections.CtorFromCollection<Int32>
| Benchmark | Baseline | Test | Test/Base | Test Quality | Edge Detector | Baseline IR | Compare IR | IR Ratio |
|---|---|---|---|---|---|---|---|---|
| 25.07 μs | 26.45 μs | 1.05 | 0.02 | False |
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.Collections.CtorFromCollection<Int32>*'Details
System.Collections.CtorFromCollection<Int32>.ConcurrentDictionary(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 | Windows 10.0.22621 |
| Queue | TigerWindows |
| Baseline | 30b34e6d7b475c1de91623d3589723181a260523 |
| Compare | 3e572e9d09026f386a57302bfd624abd7b660fb5 |
| Diff | Diff |
| Configs | CompilationMode:tiered, RunKind:micro |
Regressions in System.Collections.TryAddGiventSize<Int32>
| Benchmark | Baseline | Test | Test/Base | Test Quality | Edge Detector | Baseline IR | Compare IR | IR Ratio |
|---|---|---|---|---|---|---|---|---|
| 4.32 μs | 6.30 μs | 1.46 | 0.02 | False |
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.Collections.TryAddGiventSize<Int32>*'Details
System.Collections.TryAddGiventSize<Int32>.Dictionary(Count: 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 | Windows 10.0.22621 |
| Queue | TigerWindows |
| Baseline | 30b34e6d7b475c1de91623d3589723181a260523 |
| Compare | 3e572e9d09026f386a57302bfd624abd7b660fb5 |
| 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 |
|---|---|---|---|---|---|---|---|---|
| 2.80 μs | 6.96 μs | 2.48 | 0.08 | True |
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.Collections.TryGetValueFalse<Int32, Int32>*'Details
System.Collections.TryGetValueFalse<Int32, Int32>.Dictionary(Size: 512)
ETL Files
Histogram
JIT Disasms
Docs
Profiling workflow for dotnet/runtime repository
Benchmarking workflow for dotnet/runtime repository
Activity
LoopedBard3 commentedon Aug 6, 2024
Related Concurrent dictionary regressions: dotnet/perf-autofiling-issues#39502, dotnet/perf-autofiling-issues#39463
Mono related regressions: dotnet/perf-autofiling-issues#39525
[-][Perf] Windows/x64: 18 Regressions on 8/2/2024 1:16:56 PM[/-][+][Perf] Windows/x64: Regressions in System.Collections on 8/2/2024 1:16:56 PM[/+]LoopedBard3 commentedon Aug 6, 2024
Likely caused by: #105380
Comparison range: e60db07...fd99ad5
dotnet-policy-service commentedon Aug 6, 2024
Tagging subscribers to this area: @dotnet/area-system-collections
See info in area-owners.md if you want to be subscribed.
stephentoub commentedon Aug 6, 2024
It's possible my change impacted ConcurrentDictionary, but it wouldn't have had any effect on Dictionary, which according to the above graph is showing a huge regression.
stephentoub commentedon Aug 8, 2024
Re-opening to help track that the regressions are indeed fixed. The Dictionary one is very likely something else.
eiriktsarpalis commentedon Aug 12, 2024
I see the numbers have recovered in all three benchmarks, so we can close this.