A C# console program to microbenchmark equality methods. It confirms that there is NO performance difference with `ReferenceEquals` and `==` operator for reference type (class).

EqualityBenchmark.cs

using BenchmarkDotNet.Attributes;

public class EqualityBenchmark
{
    private const int NumberOfComparisons = 100000;

    [Benchmark]
    public string EqualityWithOperator()
    {
        var a = new Foo();
        var b = new Foo();
        var counter = 0;
        for (int i = 0; i < NumberOfComparisons; i++)
        {
            if (a == b)
            {
                counter++;
            }
        }

        return counter.ToString();
    }

    [Benchmark]
    public string EqualityWithReferenceEquals()
    {
        var a = new Foo();
        var b = new Foo();
        var counter = 0;
        for (int i = 0; i < NumberOfComparisons; i++)
        {
            if (ReferenceEquals(a, b))
            {
                counter++;
            }
        }

        return counter.ToString();
    }

    internal class Foo
    {
    }
}

Program.cs

using BenchmarkDotNet.Running;

internal class Program
{
    private static void Main(string[] args)
    {
        BenchmarkRunner.Run<EqualityBenchmark>();
    }
}

Results.txt

|                      Method |     Mean |    Error |   StdDev |
|---------------------------- |---------:|---------:|---------:|
|        EqualityWithOperator | 46.73 us | 0.023 us | 0.019 us |
| EqualityWithReferenceEquals | 46.73 us | 0.007 us | 0.005 us |