You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, our CI script is basically completely serialized: all the
steps that we perform on a given toolchain run after each other. That
has a negative impact on turn around time, because the checks that fail
most frequently (due to small nits pointed out by clippy or a format
violation detected by rustfmt) are run pretty much last. Essentially, we
don't fail fast.
As a first step towards improving the situation, this change makes the
cargo-doc check a separate job that can run in parallel to everything
else. It means that we now only check docs on the most recent stable
toolchain, but that's entirely sufficient.
Signed-off-by: Daniel Müller <deso@posteo.net>
0 commit comments