Experiment: make cross-thread run/status flags std::atomic (#3798 batches A/B)

[mcfnord]

This follows up on the discussion in #3786 and covers batches A and B of the audit in #3798 — the thread-lifecycle bRun flags and the jitter-buffer status flags. Offering it as an experiment: it is the least invasive slice of #3798, intended to establish the std::atomic pattern in the codebase and let us observe it through CI and normal use before deciding whether the more involved findings (C–F) are worth touching.

What changes

Five members become std::atomic<bool>, keeping plain operator syntax at every call site (per the conclusion in #3786 that the overloaded operators are simplest, and the default sequentially-consistent operations cost nothing at these rates):

Member	Written by	Read by
CHighPrecisionTimer::bRun (Mac/Linux variant)	main thread (Start()/Stop())	timer thread loop; socket thread via IsRunning()
CSoundBase::bRun	main thread (Start()/Stop())	audio callback thread (JACK/CoreAudio/Oboe/ASIO)
CSoundBase::bCallbackEntered	audio callback thread	GUI sound-device check
CSocket::bJitterBufferOK	socket thread	GUI read-and-reset
CClient::bJitterBufferOK	sound thread	GUI read-and-reset

A sixth member, CSocketThread::bRun (the exact #3786 case), was originally part of this PR but has been dropped: #3788 already converts it — and removes CSocket::Close() in favour of non-blocking reads, which also disposes of finding F1 in #3798. The <atomic> include added to socket.h here matches #3788's placement so the two branches merge cleanly in either order.

The only non-declaration change: the two GetAndResetbJitterBufferOKFlag() functions now use exchange ( true ) instead of a separate read and reset. Behavior is the same, and it closes a small lost-update window — a "not OK" written by the socket/sound thread between the read and the reset used to be dropped silently.

Behavior

No intended behavior change. On x86 these compile to essentially the same instructions; the difference is that the compiler may no longer cache the flag across loop iterations, and on weakly-ordered CPUs (Apple Silicon, Android, ARM Linux servers) the cross-thread store is guaranteed to become visible and correctly ordered.

Tested

• Full Linux client+server build (Qt 5.15, JACK enabled): compiles with no new warnings.
• Headless server smoke test: starts, runs, exits cleanly on SIGTERM — this exercises the modified Stop() paths of the socket and timer threads.
• macOS / Windows / Android builds not tested locally (relying on CI); the touched accesses in those backends are plain reads/writes covered by the operator overloads.

🤖 Generated with Claude Code

[softins]

Note that this one is already addressed as part of #3788, which is just awaiting a second approval.

[mcfnord]

Thanks — dropped from this PR in favour of #3788 (kept the bJitterBufferOK conversion in the same file, and matched your <atomic> include placement so the branches merge cleanly in either order). Nice bonus: removing CSocket::Close() also disposes of finding F1 in #3798.

[ann0see]

Also you need to fix the coding style violations

[mcfnord]

Also you need to fix the coding style violations

This error class should be prevented when help about the formatting rules (or a link to the rules) appears in AGENTS.md. I don't see your draft of AGENTS.md as a PR, and I think you said you'd be providing one, so I remember yesterday my agent linked to my draft in a comment. For now, my CLAUDE.md has added the following guidance, including the note that eventually this will reach the AGENTS.md file when it exists:

Upstream PRs (jamulussoftware/jamulus)

Upstream CONTRIBUTING.md is not auto-loaded context — read it before starting any upstream-bound change. (This section migrates to upstream AGENTS.md when that lands; the draft on the agent-deploy-docs branch already carries these rules, and PR #3793 adds the CONTRIBUTING.md pointer to it.)

Style, before every commit: clang-format-14 -i on all touched .cpp/.h/.mm (installed locally 2026-07-18; CI pins clang-format 14 in .github/workflows/coding-style-check.yml). make clang_format is the repo target but invokes unversioned clang-format, which this box doesn't have. Beware AlignConsecutiveDeclarations: widening one declaration's type re-pads its neighbors — this cost PR #3800 a red X.

• Qt ≥ 5.12.2 APIs only (QT_VERSION_CHECK guards); C++11; smallest-possible single-topic PRs; features agreed in an issue/Discussion before coding.
• Working pattern (issue Audit: non-atomic booleans and ints shared across threads (follow-up to #3786) #3798 → PR Experiment: make cross-thread run/status flags std::atomic (#3798 batches A/B) #3800): worktree off freshly fetched main in the scratchpad (never build upstream code in /root/jamulus — it clobbers the deploy-cached ./Jamulus), full build + headless SIGTERM smoke test, push branch to fork, gh pr create --repo jamulussoftware/jamulus --head mcfnord:<branch> --draft.

[softins]

This is nice, and more concise than the original!