Fix core sync test deadlock on darwin #4253
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Feoramund
reviewed
Sep 16, 2024
|
Great! I wonder, if you do -minimum-os-version:14.4.0 (so it uses the new public os_sync api) does that also not deadlock? I assume it doesn't? |
Co-authored-by: Feoramund <[email protected]>
|
And indeed, would want an improvement like this in the compiler too, but if you aren't comfortable there I can pick that up too. |
|
Can confirm no deadlocks with |
|
Another thing catching my eye (clicking through that link) is that ulock_wait is microseconds and ulock_wait2 is nanoseconds. Looks like we handle that wrong here too, passing in nanoseconds where the api expects microseconds. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The deadlock was discussed in #4232.
I'm running macos 14.6.1 and
__ulock_waitrandomly deadlocks in the core sync tests. I guess it shouldn't be too surprising since this isn't really a public API and as the zig guys discovered Apple has been migrating to__ulock_wait2.In any case using
__ulock_wait2seems to fix the problem. The corresponding change should probably be made to the C++ code as well.