EgorBot for EgorBo in #109014

EgorBot commented 1 month ago

Processing https://github.com/dotnet/runtime/pull/109014#issuecomment-2422362712 command:

Command

-intel -arm64 -profiler -nonativepgo ```cs using BenchmarkDotNet.Attributes; using BenchmarkDotNet.Running; BenchmarkSwitcher.FromAssembly(typeof(Bench).Assembly).Run(args); [GcServer(true)] public class Bench { static object[] Src0 = Enumerable.Range(0, 2).Select(i => (object)i).ToArray(); static object[] Src1 = Enumerable.Range(0, 10).Select(i => (object)i).ToArray(); static object[] Src2 = Enumerable.Range(0, 128).Select(i => (object)i).ToArray(); static object[] Dst = new object[128]; [Benchmark]public void CopyObjects2() => Src0.AsSpan().CopyTo(Dst); [Benchmark]public void CopyObjects10() => Src1.AsSpan().CopyTo(Dst); [Benchmark]public void CopyObjects128() => Src2.AsSpan().CopyTo(Dst); } ```

(EgorBot will reply in this issue)

EgorBot commented 1 month ago

Benchmark results on `Arm64`

BenchmarkDotNet v0.14.0, Ubuntu 24.04 LTS (Noble Numbat)
Arm64
  Job-IUCNRB : .NET 10.0.0 (42.42.42.42424), Arm64 RyuJIT AdvSIMD
  Job-YJVBKF : .NET 10.0.0 (42.42.42.42424), Arm64 RyuJIT AdvSIMD
Server=True

Method	Toolchain	Mean	Error	Ratio
CopyObjects2	Main	8.093 ns	0.0013 ns	1.00
CopyObjects2	PR	7.724 ns	0.0014 ns	0.95

CopyObjects10	Main	10.611 ns	0.0023 ns	1.00
CopyObjects10	PR	9.729 ns	0.0017 ns	0.92

CopyObjects128	Main	23.811 ns	0.0111 ns	1.00
CopyObjects128	PR	51.363 ns	0.1233 ns	2.16

BDN_Artifacts.zip

Profile for `Bench_CopyObjects128`:

Flame graphs: Main vs PR 🔥 Speedscope: Main vs PR Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR

Profile for `Bench_CopyObjects2`:

Flame graphs: Main vs PR 🔥 Speedscope: Main vs PR Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR

Profile for `Bench_CopyObjects10`:

Flame graphs: Main vs PR 🔥 Speedscope: Main vs PR Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR

EgorBot commented 1 month ago

cc @EgorBo (logs)

EgorBot commented 1 month ago

Benchmark results on `Intel`

BenchmarkDotNet v0.14.0, Ubuntu 24.04 LTS (Noble Numbat)
Intel Xeon Platinum 8488C, 1 CPU, 16 logical and 8 physical cores
  Job-DBRMHF : .NET 10.0.0 (42.42.42.42424), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
  Job-YPTGEU : .NET 10.0.0 (42.42.42.42424), X64 RyuJIT AVX-512F+CD+BW+DQ+VL+VBMI
Server=True

Method	Toolchain	Mean	Error	Ratio
CopyObjects2	Main	8.438 ns	0.0325 ns	1.00
CopyObjects2	PR	6.560 ns	0.0429 ns	0.78

CopyObjects10	Main	8.862 ns	0.0520 ns	1.00
CopyObjects10	PR	8.021 ns	0.0712 ns	0.91

CopyObjects128	Main	18.749 ns	0.0324 ns	1.00
CopyObjects128	PR	26.767 ns	0.0521 ns	1.43

BDN_Artifacts.zip

Profile for `Bench_CopyObjects128`:

Flame graphs: Main vs PR 🔥 Speedscope: Main vs PR Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR

Profile for `Bench_CopyObjects2`:

Flame graphs: Main vs PR 🔥 Speedscope: Main vs PR Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR

Profile for `Bench_CopyObjects10`:

Flame graphs: Main vs PR 🔥 Speedscope: Main vs PR Hot asm: Main vs PR Hot functions: Main vs PR Counters: Main vs PR

EgorBot commented 1 month ago

cc @EgorBo (logs)

EgorBot / runtime-utils