Release v0.5.8 (httpz epoll recv-on-freed-Conn fix)#123
Conversation
|
Warning Review limit reached
Next review available in: 30 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Fixes #120: a recurring SIGSEGV during normal operation under connection churn.
http.zig's epoll worker closed a finished connection's socket on a worker thread (to implicitly drop it from epoll), then recycled the Conn later via processSignal. That close raced epoll_wait, so the fd could stay armed past the recycle and a later batch delivered a
.recvfor freed memory ->getStateon a null/recycled HTTPConn (fault at the_stateoffset). Confirmed reproducing on v0.5.7 every few hours.The fix removes the fd from epoll and closes it on the loop thread, before disown recycles the Conn (mirroring the synchronous EPOLL_CTL_DEL that HTTPConn.disown already does for res.disown()).
Pinned to a temporary privkeyio http.zig fork (
0c6cd82, = upstream8dc6441+ the one fix). Upstream PR karlseguin/http.zig#216; will repoint to upstream and drop the fork once it merges.0c6cd82Verified: http.zig 125-test suite passes; wisp unit tests pass; survived 4000 connect/GET/close + partial-close churn cycles with no crash, then clean shutdown.