dotnet / perf-autofiling-issues

A landing place for auto-filed performance issues before they receive triage
MIT License
9 stars 4 forks source link

[Perf] Linux/x64: 9 Improvements on 7/12/2024 11:27:13 PM #38635

Open performanceautofiler[bot] opened 4 months ago

performanceautofiler[bot] commented 4 months ago

Run Information

Name Value
Architecture x64
OS ubuntu 22.04
Queue TigerUbuntu
Baseline b8ac8b2d7f79aec93f9cad722f3cc4ccb931b826
Compare ac0eeb015a0266adb06696bb58c21e5bfe6d9b56
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in System.Tests.Perf_Random

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
101.69 ns 74.36 ns 0.73 0.32 False
75.25 ns 55.75 ns 0.74 0.43 False
74.44 ns 55.93 ns 0.75 0.39 False
79.83 ns 61.89 ns 0.78 0.40 False
107.09 ns 82.69 ns 0.77 0.36 False
99.58 ns 77.30 ns 0.78 0.32 False

graph graph graph graph graph graph 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.Tests.Perf_Random*'
### System.Tests.Perf_Random.Next #### ETL Files #### Histogram #### JIT Disasms ### System.Tests.Perf_Random.Next_unseeded #### ETL Files #### Histogram #### JIT Disasms ### System.Tests.Perf_Random.NextSingle_unseeded #### ETL Files #### Histogram #### JIT Disasms ### System.Tests.Perf_Random.NextDouble_unseeded #### ETL Files #### Histogram #### JIT Disasms ### System.Tests.Perf_Random.NextSingle #### ETL Files #### Histogram #### JIT Disasms ### System.Tests.Perf_Random.NextDouble #### ETL Files #### Histogram #### JIT Disasms ### Docs [Profiling workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/profiling-workflow-dotnet-runtime.md) [Benchmarking workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/benchmarking-workflow-dotnet-runtime.md)

Run Information

Name Value
Architecture x64
OS ubuntu 22.04
Queue TigerUbuntu
Baseline b8ac8b2d7f79aec93f9cad722f3cc4ccb931b826
Compare ac0eeb015a0266adb06696bb58c21e5bfe6d9b56
Diff Diff
Configs CompilationMode:wasm, RunKind:micro

Improvements in PerfLabTests.CastingPerf

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
452.90 ns 368.81 ns 0.81 0.07 True
175.93 ns 56.83 ns 0.32 0.10 True
178.98 ns 67.11 ns 0.37 0.12 True

graph graph graph 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 'PerfLabTests.CastingPerf*'
### PerfLabTests.CastingPerf.CheckArrayIsVariantGenericInterfaceReflection #### ETL Files #### Histogram #### JIT Disasms ### PerfLabTests.CastingPerf.CheckListIsVariantGenericInterface #### ETL Files #### Histogram #### JIT Disasms ### PerfLabTests.CastingPerf.AssignArrayElementByVariance #### ETL Files #### Histogram #### JIT Disasms ### Docs [Profiling workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/profiling-workflow-dotnet-runtime.md) [Benchmarking workflow for dotnet/runtime repository](https://github.com/dotnet/performance/blob/master/docs/benchmarking-workflow-dotnet-runtime.md)
lewing commented 3 months ago

the casting perf improvements are from https://github.com/dotnet/runtime/pull/103757 the others appear to be noise