BenchmarkDotNet
BenchmarkDotNet copied to clipboard
Powerful .NET library for benchmarking
This is a prototype for supporting any type of awaitable, stemmed from the discussion in #2111. (It might even be able to supersede that PR, since I am seeing even...
Followup to #2309. Overhead matches workload return type. Use Consumer instead of writing to generic field. Added support for pointer, ByRef, and ValueTask returns.
This PR adds an option to run toolchain generator and builder without actually executing the built binary. This way we can attach other tools such as perfview and Visual Studio...
fix #1715
Before #2148, like v0.13.2:  Master:  PR:  ## no validation errors: Master:  PR: 
as discussed in #1967 cc @adamsitnik
Fixes #1164 Fixes #2022 Look at "Update existing tests" and "Update new tests" to see what has changed. # Fixes: ## https://github.com/dotnet/BenchmarkDotNet/issues/1164#issue-448217544 Master: ``` | Type | Method | Mean...
closes [#190](https://github.com/dotnet/BenchmarkDotNet/issues/190)
Null value is ok, but it is shown as unknown. It's impossible to get Unknown value now, but I decided to add handling for this situation.
The 3rd commit shows what's changed. I removed `Summary.IsMultipleRuntimes` because it is actually redundant. Now the `Toolchain` column will be showed only when using `.WithToolchain()` method. ```C# Job.Default.WithRuntime(Net60); // toolchain...