app-server: forwarded core events rebuild the subscriber list

What variant of Codex are you using?

App Server

What feature would you like to see?

I found a potential performance concern in current app-server source rather than a confirmed user-visible bug. Please consider retaining an immutable subscriber snapshot for a listener and replacing it only when thread membership changes, if that can preserve the current subscription and teardown semantics.

Additional information

Evidence

• Source-proven: thread_lifecycle.rs:309-340 calls subscribed_connection_ids(conversation_id).await for every forwarded core event before constructing the thread-scoped sender.
• Source-proven: thread_state.rs:308-323 stores all thread entries behind one Arc<Mutex<ThreadStateManagerInner>>; subscribed_connection_ids acquires that mutex and collects the thread's HashSet into a fresh Vec.
• Source-proven: membership is mutated by explicit subscribe, unsubscribe, add, and connection-removal paths in thread_state.rs:472-589, whereas the lookup above runs per forwarded event.
• Historical context only: closed, unmerged PR #29545 proposed a watcher-backed immutable connection-ID snapshot, including an empty snapshot at teardown. Its bundled benchmark claims are not measurement for this report, and the audited current source still rebuilds the vector.

Impact

Not measured. Global-lock contention, allocation volume, and listener latency have not been measured. Each forwarded event awaits a mutex shared by the manager's thread map and allocates a vector containing all current subscriber IDs; the work may therefore grow with event rate, subscriber count, and concurrent manager activity. No production impact has been observed.

Question

Would it make sense to retain one immutable connection-ID snapshot per listener and publish a replacement only on membership mutations, while publishing an empty snapshot at teardown so in-flight delivery cannot address stale connections?

I checked all relevant issues, comments, pull requests, discussions, and release notes; this report is not a duplicate.

I am reporting this finding only and am not proposing a pull request unless a maintainer invites one.

Disclosure

Investigated thoroughly with GPT-5.6 Sol (runtime-default reasoning effort), using Oh My Pi as the agent framework.

This report is not generic or unreviewed AI-generated output. Its claims were checked against the cited evidence, and it includes the relevant detail intended to help maintainers resolve the issue.

If reports like this are not useful to the project, please let me know and I will refrain from submitting similar ones. My intent is to help without wasting maintainer time or energy or discouraging their work.

Thank you for your work.