- [The results site](#the-results-site)

The results site (https://fastruby.github.io/fast-ruby/) compares Rubies, so every build of a benchmark has to run on the same machine.

`script/run_cross_ruby.rb` (Ruby 3 on your machine, it calls Docker) does that, and runs the newest released MRI again after every 3 builds, so the site can show how steady the machine was.

To see a few benchmarks the way the site shows them, run them on a few builds, build the site into `_site/`, then open `_site/index.html` in a browser:

`ruby script/run_cross_ruby.rb --help` lists the options.

The builds are the ones in the CI matrix (`.github/workflows/benchmarks.yml`), so a new Ruby added there and in `compose.yaml` is measured too, and once released it becomes both the reference and the Ruby the site opens on.

On an Apple silicon Mac, leave out `ruby_2.1` and `jruby_9.1` (they run under emulation, so their numbers are not comparable) and `ruby_3.1+yjit` (Ruby 3.1's YJIT only exists on x86-64).

CI has two workflows:

- `.github/workflows/benchmarks.yml` checks that every benchmark runs on every Ruby: on a PR, the benchmark files it changes; on `main`, all of them when a benchmark or a shared file changed. It publishes nothing.

- `.github/workflows/results-site.yml` runs every build on every file once a week, split over 6 machines, and publishes the site. Run it by hand from the Actions tab to publish sooner.