dcpomatic/src/lib/encode_server_finder.cc, branch v2.16.18DCP-o-matic DCP toolshttps://git.carlh.net/cgit/dcpomatic/atom?h=v2.16.182022-02-25T07:00:51ZAvoid boost::bind()ing a shared_ptr.2022-02-25T07:00:51ZCarl Hetheringtoncth@carlh.net2022-02-22T19:34:08Zurn:sha1:b2c412480aa2cc29bfbb90556f974e9294608898Fix some dubious thread/locking behaviour.2022-02-25T07:00:51ZCarl Hetheringtoncth@carlh.net2022-02-24T18:47:19Zurn:sha1:5ec4efca61906608e2e7a590b3d1c3f2b7078e81
Previously we had server_found(), which took the lock and found
a server, which it returned as an iterator into the list.
However, it then released the lock, which I think left the
iterator unprotected.
This wasn't done in response to any particular bug, I just
noticed it on the way past.
C++11 tidying.2021-05-03T18:18:04ZCarl Hetheringtoncth@carlh.net2021-05-03T09:17:34Zurn:sha1:689fa55d1529ad88449ca464e9107c4dcc54d1cbAdd some possibly-useful markers for debugging threads from coredumps.2021-04-20T22:52:07ZCarl Hetheringtoncth@carlh.net2021-04-20T10:27:40Zurn:sha1:c370c651eba466f5073384de8b304a2625c64b89BOOST_FOREACH.2021-01-07T22:27:56ZCarl Hetheringtoncth@carlh.net2021-01-04T20:32:13Zurn:sha1:a5d004b0773f633401528392fc28e66d70e13ac8std::shared_ptr2021-01-07T21:48:29ZCarl Hetheringtoncth@carlh.net2021-01-04T20:16:53Zurn:sha1:dd9be86db6cde0afa5da0d1d1ac43b42e05dca26If we don't query a server (because we already know about it)2020-11-26T01:24:32ZCarl Hetheringtoncth@carlh.net2020-11-26T01:05:25Zurn:sha1:dfb23c675ca145b5570ccb5790ea804ad736a247
the "last seen time" will never be updated, so the server will
be discarded.
It seems that we should always ping servers (so that set_seen gets
called on receipt of the response), no matter whether
"auto-discovered" or configured, so that the "discard" code doesn't
kick in.
Otherwise we remove and re-add our configured servers every
10 seconds, which is inefficient and which possibly triggers
other bugs.
Use a foreach.2020-11-26T00:49:19ZCarl Hetheringtoncth@carlh.net2020-11-26T00:22:13Zurn:sha1:6e5dec71bf4e7a5ff81ecc115c04e8ec2b540c67Build fixes for Boost >= 1.732020-09-01T14:34:26ZCarl Hetheringtoncth@carlh.net2020-08-27T20:37:32Zurn:sha1:c04fec82d25127fafa73c3daff87bece9aa8c8e8
Forward-ported-from: d1e9749ca290673639a49d693a8fe5c6557cc2de
Fix bugs in thread termination causing occasional pthread2020-07-29T18:22:54ZCarl Hetheringtoncth@carlh.net2020-07-29T18:22:54Zurn:sha1:e3c7656f9dc0acbaf518c051b847ee2e4eb7ba23
assertion failures.
Before this, it was possible for J2KEncoder::terminate_threads()
to finish without terminating all threads if the thread _running_
terminate_threads() was itself interrupt()ed.
This is because the thread_group::join_all() in terminate_threads()
is an interruption point, so it was possible it not to complete
but instead to throw interrupted_exception. Then the owning
J2KEncoder would be torn down but the threads would still be running,
causing use-after-frees.
This commit adds some boost::this_thread::disable_interruption
objects to ensure that the owning thread is not interrupted while
it is being destroyed.
Also tidy up code that does this stuff, assuming that it's safe
to not call thread::joinable but instead do
thread.interrupt();
try {
thread.join();
} catch (...) {}