Julia v1.0.0 used for Micro-Benchmarks?

[MagicMuscleMan]

Is your feature request related to a problem? Please describe.
On the Julia Mcro-Benchmarks page it is stated, that the benchmarks show Julia v1.0.0 results. However, in the Microbenchmark repository currently Julia v1.9.3 seems to be used.

Describe the solution you'd like
Either the version numbers on the benchmark page should be updated (ideally in a way to always represent the correct versions in the future) or the benchmark should be repeated with a more up-to-date Julia version.

[ViralBShah]

Maybe someone can submit a PR to the benchmarks file updating everything and seeing how it all works out: https://github.com/JuliaLang/Microbenchmarks/blob/master/.github/workflows/benchmarks.yml

[MagicMuscleMan]

PR created. Note, that I don't know whether anything of this will work.

Even if it should work, there is still the Julia Micro-Benchmark page stating that Julia v1.0.0 is used and I do not know whether the text is incorrect or whether the figure never has been updated since v1.0.0.

[ViralBShah]

That page has not been updated since 1.0.

[MagicMuscleMan]

PR seems to build without error after updating some stuff and deactivating Go. Could you please go through it and work towards approving it? Without the PR this is in a non-useful state if I understand this correctly.

If there is someone with Go knowledge, they can reactivate the Go benchmarks later.

If the benchmark really works (I think I saw some numbers where the whole result might have been calculated during compilation), could you then also update the web page to show 1.11 numbers? As long as we do not recommend Julia 1.0 to the user, we should not show the 1.0 numbers as the official Julia performance values.

[ViralBShah]

Used the JuliaCon hackathon to clean up the repo and generated a new chart. PR: JuliaLang/Microbenchmarks.jl#85

[linwaytin]

Awesome! What is the Julia version for this plot?

[ViralBShah]

1.11.