Open EgorBot opened 2 months ago
c7a_4xlarge
BenchmarkDotNet v0.14.0, Ubuntu 24.04 LTS (Noble Numbat)
AMD EPYC 9R14, 1 CPU, 16 logical and 16 physical cores
Job-CJHJBZ : .NET 9.0.0 (42.42.42.42424), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
Job-HJWPTH : .NET 9.0.0 (42.42.42.42424), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
EnvironmentVariables=DOTNET_JitDisasm=TrailingZeroCount_ulong
Method | Toolchain | Mean | Error | Ratio |
---|---|---|---|---|
TrailingZeroCount_ulong | Before | 279.7 ns | 0.11 ns | 1.00 |
TrailingZeroCount_ulong | After | 279.8 ns | 0.08 ns | 1.00 |
Flame graphs: Main vs PR 🔥 Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR
For clean perf
results, make sure you have just one [Benchmark]
in your app.
Amd
BenchmarkDotNet v0.14.0, Ubuntu 22.04.4 LTS (Jammy Jellyfish)
AMD EPYC 7763, 1 CPU, 16 logical and 8 physical cores
Job-WVOZHL : .NET 9.0.0 (42.42.42.42424), X64 RyuJIT AVX2
Job-PYVXXX : .NET 9.0.0 (42.42.42.42424), X64 RyuJIT AVX2
EnvironmentVariables=DOTNET_JitDisasm=TrailingZeroCount_ulong
Method | Toolchain | Mean | Error | Ratio |
---|---|---|---|---|
TrailingZeroCount_ulong | Before | 624.8 ns | 0.66 ns | 1.00 |
TrailingZeroCount_ulong | After | 624.6 ns | 0.11 ns | 1.00 |
Flame graphs: Main vs PR 🔥 Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR
For clean perf
results, make sure you have just one [Benchmark]
in your app.
Processing https://github.com/dotnet/runtime/issues/107304#issuecomment-2331870986 command:
Command
-amd -awsamd -commit 72f9ee0d26c23b3c58ec5af8aeea316095761646 vs previous -profiler --envvars DOTNET_JitDisasm:TrailingZeroCount_ulong ```cs using BenchmarkDotNet.Attributes; using BenchmarkDotNet.Running; using System.Numerics; using System.Runtime.CompilerServices; public class Perf_BitOperations { static T[] Array(EgorBot will reply in this issue)