Fix unsafe usage of Base.Condition #234
Conversation
|
@krynju @ExpandingMan please give this a try |
jpsamaroo
force-pushed
the
jps/thread-safe-notify
branch
from
d324987 to
a7a06b9
Compare
|
Works fine for me! |
|
Looks good to me as well, thanks for the quick fix. Btw, I'm pretty sure this problem exists on your recent tag, so a new version tag is likely in order. Also, this is making me nervous about tests, which don't seem to have been running. Is your intention for master to always be passing? We should probably figure out why tests aren't running. If you'd like I can do a PR to run github workflow tests, though I'm not sure how nicely they play with multiple processes. |
I just submitted 0.11.5, if that's what you mean.
The Buildkite tests are the most important, and they're generally passing (with some stochastic failures on nightly that I don't have the bandwidth to debug right now). Windows is just very unreliable with Dagger testing, but it is good to have a sense for how well things are working for those users.
If possible, yes, but I can tolerate some failures due to random chance. Much of the unreliability in CI we see appears to be due to Distributed, which I haven't had time to sit down and figure out. |
|
Maybe this is a stupid question but are the buildkite tests visible somewhere? I don't think I'm seeing them in the PR's... |
|
Oh, shoot, you're supposed to be able to see them. @DilumAluthge any ideas? |
No description provided.