KEMBAR78
[Perf] Windows/arm64: 16 Regressions on 6/5/2024 10:36:16 PM · Issue #36252 · dotnet/perf-autofiling-issues · GitHub
Skip to content

[Perf] Windows/arm64: 16 Regressions on 6/5/2024 10:36:16 PM #36252

@performanceautofiler

Description

@performanceautofiler

Run Information

Name Value
Architecture arm64
OS Windows 10.0.22621
Queue AmpereWindows
Baseline 9a3d45a66d96dd73b360471e0de8f0c5fb7c89c0
Compare 60edbf9371ff00349e9224fb0920c8bca51f1a4f
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Single>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
33.22 ns 65.89 ns 1.98 0.72 True
44.55 ns 97.19 ns 2.18 0.75 True
32.29 ns 38.79 ns 1.20 0.74 False
112.56 ns 172.11 ns 1.53 0.73 False

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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives&lt;Single&gt;*'

System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Single>.Max_Scalar(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Single>.Max_Vector(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Single>.Abs(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Single>.MaxMagnitude_Scalar(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

Docs

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


Run Information

Name Value
Architecture arm64
OS Windows 10.0.22621
Queue AmpereWindows
Baseline 9a3d45a66d96dd73b360471e0de8f0c5fb7c89c0
Compare 60edbf9371ff00349e9224fb0920c8bca51f1a4f
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
34.27 ns 42.03 ns 1.23 0.72 False
111.78 ns 154.69 ns 1.38 0.68 False
115.69 ns 175.95 ns 1.52 0.70 False

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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives&lt;Int32&gt;*'

System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.Add_Scalar(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.MaxMagnitude_Scalar(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_NumberTensorPrimitives<Int32>.MaxMagnitude_Vector(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

Docs

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


Run Information

Name Value
Architecture arm64
OS Windows 10.0.22621
Queue AmpereWindows
Baseline 9a3d45a66d96dd73b360471e0de8f0c5fb7c89c0
Compare 60edbf9371ff00349e9224fb0920c8bca51f1a4f
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int64>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
0.91 ns 3.86 ns 4.24 0.93 False

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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Runtime.Intrinsics.Tests.Perf_Vector128Of&lt;Int64&gt;*'

System.Runtime.Intrinsics.Tests.Perf_Vector128Of<Int64>.DotBenchmark

ETL Files

Histogram

JIT Disasms

Docs

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


Run Information

Name Value
Architecture arm64
OS Windows 10.0.22621
Queue AmpereWindows
Baseline 9a3d45a66d96dd73b360471e0de8f0c5fb7c89c0
Compare 60edbf9371ff00349e9224fb0920c8bca51f1a4f
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives<Single>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
34.96 ns 52.85 ns 1.51 0.74 False

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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives&lt;Single&gt;*'

System.Numerics.Tensors.Tests.Perf_FloatingPointTensorPrimitives<Single>.Distance(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

Docs

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


Run Information

Name Value
Architecture arm64
OS Windows 10.0.22621
Queue AmpereWindows
Baseline 9a3d45a66d96dd73b360471e0de8f0c5fb7c89c0
Compare 60edbf9371ff00349e9224fb0920c8bca51f1a4f
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Numerics.Tensors.Tests.Perf_BinaryIntegerTensorPrimitives<Byte>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
20.62 ns 64.33 ns 3.12 0.77 True
134.97 ns 148.42 ns 1.10 0.76 False
105.06 ns 155.64 ns 1.48 0.72 False
14.53 ns 45.99 ns 3.17 0.66 True
12.43 ns 17.58 ns 1.41 0.74 False

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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tensors.Tests.Perf_BinaryIntegerTensorPrimitives&lt;Byte&gt;*'

System.Numerics.Tensors.Tests.Perf_BinaryIntegerTensorPrimitives<Byte>.TrailingZeroCount(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_BinaryIntegerTensorPrimitives<Byte>.TrailingZeroCount(BufferLength: 3079)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_BinaryIntegerTensorPrimitives<Byte>.PopCount(BufferLength: 3079)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_BinaryIntegerTensorPrimitives<Byte>.PopCount(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

System.Numerics.Tensors.Tests.Perf_BinaryIntegerTensorPrimitives<Byte>.LeadingZeroCount(BufferLength: 128)

ETL Files

Histogram

JIT Disasms

Docs

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


Run Information

Name Value
Architecture arm64
OS Windows 10.0.22621
Queue AmpereWindows
Baseline 9a3d45a66d96dd73b360471e0de8f0c5fb7c89c0
Compare 60edbf9371ff00349e9224fb0920c8bca51f1a4f
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Numerics.Tests.Perf_Matrix4x4

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
1.65 ns 6.74 ns 4.09 0.89 False

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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_Matrix4x4*'

System.Numerics.Tests.Perf_Matrix4x4.Transpose

ETL Files

Histogram

JIT Disasms

Docs

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


Run Information

Name Value
Architecture arm64
OS Windows 10.0.22621
Queue AmpereWindows
Baseline 9a3d45a66d96dd73b360471e0de8f0c5fb7c89c0
Compare 60edbf9371ff00349e9224fb0920c8bca51f1a4f
Diff Diff
Configs CompilationMode:tiered, RunKind:micro

Regressions in System.Numerics.Tests.Perf_VectorOf<UInt64>

Benchmark Baseline Test Test/Base Test Quality Edge Detector Baseline IR Compare IR IR Ratio
0.99 ns 2.21 ns 2.23 0.90 False

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
py .\performance\scripts\benchmarks_ci.py -f net8.0 --filter 'System.Numerics.Tests.Perf_VectorOf&lt;UInt64&gt;*'

System.Numerics.Tests.Perf_VectorOf<UInt64>.DotBenchmark

ETL Files

Histogram

JIT Disasms

Docs

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

Metadata

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions