utils: report failure to set a thread name, and fall back to short names

[lhoward]

On Linux, pthread_setname_np fails with ERANGE for a name longer than 15 characters, and setCurrentThreadName ignores the result and returns true. Every name the library sets is longer than that (avdecc::watchDog is 16 characters, avdecc::PCapInterface::Capture 30), so none of them take effect: each thread keeps the name it inherited from the thread that created it, which makes the library's threads hard to tell apart in top, perf or a debugger.

• Return value: setCurrentThreadName now returns whether the name was set, from the result of pthread_setname_np (Linux and macOS) or prctl, and no longer truncates.
• Callers: each call site that can run on Linux falls back to a short name when the full one is rejected: avdecc:watchDog, avdecc:SM, avdecc:ctrlSM, avdecc:pcap, avdecc:serial, avdecc:local and avdecc:virtual. An executor falls back to Exec: followed by the first 10 characters of the last :: component of its name, which tells executors apart.

This overlaps #204 in src/watchDog.cpp and CHANGELOG-avdecc.md, so whichever is merged second needs a small rebase.

Testing: builds on dev, and together with the watchdog and state machine changes in #204 and #205, the unit tests give the same results as dev without them: 289 of 292 pass. INTEGRATION_EndStation.DestroyWhileMessageInflight_pcap and INTEGRATION_ProtocolInterfacePCap_F.VuDelegate fail either way, as they need pcap access to a real interface.

🤖 Generated with Claude Code

https://claude.ai/code/session_012i5Dd3gCyYqp4X7vxhSmxh

[christophe-calmejane]

I'm wondering if it wouldn't be better for the function to return the error status, and let the caller truncate itself (it probably better know how to properly reduce the length of the string instead of cutting the end, which could prevent easy identification of the thread).
Anyhow, thanks for the report/PR, I wasn't aware of this issue.

[lhoward]

Thanks Christophe, good point. I've reworked it that way: setCurrentThreadName no longer truncates, and returns whether the name was set, from pthread_setname_np's result on Linux and macOS. When the full name is rejected, each call site that can run on Linux falls back to a short name that still identifies the thread: avdecc:watchDog, avdecc:SM, avdecc:ctrlSM, avdecc:pcap, avdecc:serial, avdecc:local, avdecc:virtual, and for executors Exec: plus the last component of the executor's name. It's rebased onto dev too.

[sonarqubecloud]

Quality Gate passed

Issues
3 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud