cfallin commented on issue #4211:
Sure, added
--tests
to the CI invocation to include integration tests in all crates; it doesn't seem to cause any other issues or significantly more runtime locally (let's see if CI stays happy!).
cfallin commented on issue #4211:
Ah, actually, I have confused myself with an overly complex dev setup -- I was indeed seeing failures on my M1 Mac but didn't look closely, and they were because of a hardcoded
x86_64-unknown-linux-gnu
triple. On myx86_64-unknown-linux-gnu
system the tests pass onmain
with Rust 1.61. The issue is with Rust 1.60; i.e. CI didn't catch this because we're fine on latest stable but not version N-1.Generally our policy is that we only support latest stable, but it's a small enough tweak to a brittle error-string-matching test that I think we should still do it, and maybe it makes things a bit more robust in the future too.
Last updated: Nov 22 2024 at 16:03 UTC