Zulip Chat Archive

Stream: nightly-testing

Topic: lake is printing "build" lines even when oleans are present?


Kim Morrison (Apr 30 2024 at 11:40):

lake appears to be printing "build" lines for some initial segment of files, even when all oleans are present.

See https://github.com/leanprover-community/mathlib4/actions/runs/8892841920/job/24421612621 as an example.

Kim Morrison (Apr 30 2024 at 11:41):

It's running fast, so I think it is successfully noticing and reusing the olean. Just printing unnecessarily. One doesn't notice this on the command line because the build lines are now being overwritten.

Kim Morrison (Apr 30 2024 at 11:42):

@Mac Malone, do you see what is going on?

Ruben Van de Velde (Apr 30 2024 at 11:49):

rss-bot said:

refactor: lake: --wfail & track jobs & logs & simplify build monads (lean4#3835)

This is a major refactor of Lake's build code. The key changes:

  • Job Registration: Significant build jobs are now registered by
    build functions. The DSL inserts this registration automatically into
    user-defined targets and facets, so this change should require no
    end-user adaption. Registered jobs are incrementally awaited by the main
    build function and the progress counter now indicates how many of these
    jobs are completed and left-to-await. On the positive side, this means
    the counter is now always accurate. On the negative side, this means
    that jobs are displayed even if they are no-ops (i.e., if the target is
    already up-to-date).

Is this it? ^

Kim Morrison (Apr 30 2024 at 11:52):

Oh, yes. I didn't read that last sentence. Yes, that sounds pretty bad. It is going to be hell diagnosing people's cache problems.

Kim Morrison (Apr 30 2024 at 11:53):

I don't mind if those jobs are included in the total counter. I just don't want lines printed.


Last updated: May 02 2025 at 03:31 UTC