| Age | Commit message (Expand) | Author |
|---|
| 22 hours | Generalize the worker setup/runloop/teardown logic by adding a smol ot_worker helper | Dirk Engling |
| 27 hours | Make the dynamic accesslist readers release their line buffer and properly exit | Dirk Engling |
| 28 hours | Refactor ot_workqueue out of ot_mutex, no behaviour change | Dirk Engling |
| 31 hours | Make the mainloop behave like every worker by claiming and releasing a socket being worked on | Dirk Engling |
| 31 hours | Make tcp worker multithreaded, too | Dirk Engling |
| 46 hours | Refactor all tcp specific code into the ot_tcp module to prepare for multiple concurrent workers | Dirk Engling |
| 47 hours | Add test for new config params | Dirk Engling |
| 47 hours | Add config parsers for multiple tcp socket readers and a helper to see if we can offer it on the platform | Dirk Engling |
| 2 days | Introduce a fast path assuming that a tcp connection's payload already has arrived at accept time | Dirk Engling |
| 2 days | Start documenting design decisions and measurments | Dirk Engling |
| 2 days | Remove references that you need libowfat to build opentracker | Dirk Engling |
| 2 days | git ignore the poll flavor of the runner, too | Dirk Engling |
| 2 days | Fix idle tests in presence of an accept filter | Dirk Engling |
| 2 days | Remove stale xcode build file | Dirk Engling |
| 2 days | Fix expiry tests under the new accept filters | Dirk Engling |
| 2 days | Use accept filter dataready instead of httpready on FreeBSD | Dirk Engling |
| 2 days | Defer accepts until some data actually is there | Dirk Engling |
| 2 days | Enforce the invariant that each socket is only ever expected to be read from or written to, never both | Dirk Engling |
| 2 days | Don't treat a full send buffer as an error when sending http. Retry later | Dirk Engling |
| 3 days | Reduce syscalls and add helper to count them | Dirk Engling |
| 3 days | Add an http load generator | Dirk Engling |
| 3 days | Reduce number of per connection sys calls | Dirk Engling |
| 3 days | Check for free port in the udp integration test, also speed it up by a bit | Dirk Engling |
| 3 days | Consolidate all tcp/http tear down into a single http_close function to prevent a rare UAF for expiring connections | Dirk Engling |
| 4 days | Build everything in one sweep, without intermediate object files | Dirk Engling |
| 4 days | add wake-up tester that helped figuring out how to tickle hanging udp readers | Dirk Engling |
| 4 days | Revert last wake-up experiment. Having each udp worker wake up once a second to check if the tracker stops is wasteful | Dirk Engling |
| 4 days | Fix a test that would over-report success rate when losing the race for cycles to the test target | Dirk Engling |
| 4 days | Livesync now also gets proper deinit code | Dirk Engling |
| 4 days | With more workers blockingly waiting on udp sockets, we need to keep track of all sockets to shut them down on exit | Dirk Engling |
| 4 days | Default udp workers to 1, as it is the most sane choice everywhere, also reflect the new udp parallelism in the man pages | Dirk Engling |
| 4 days | Let the load tester log losses to better gauge performance | Dirk Engling |
| 4 days | Add load tests to sanitize check the trackerlogic, based on udp requests | Dirk Engling |
| 4 days | Performance improvement, where it works: Let different udp worker threads have their own socket | Dirk Engling |
| 5 days | The main event loop(s) now also catch an EINTR and return instead of blocking | Dirk Engling |
| 5 days | Add concurrency test | Dirk Engling |
| 5 days | Make global torrent counter atomic and fix the setter and getter | Dirk Engling |
| 5 days | Plug in the new per-thread stats counter, also make sure that the ot_clean sleep does not stall tracker exit | Dirk Engling |
| 5 days | Second half of moving the stat enum defines to trackerlogic | Dirk Engling |
| 5 days | Prepare having each thread keep their own stats to avoid data races | Dirk Engling |
| 5 days | Turn malloc+memset into calloc | Dirk Engling |
| 5 days | Some more g_now_seconds occurrences | Dirk Engling |
| 5 days | Forgot some g_now_seconds occurences | Dirk Engling |
| 5 days | All OSes now do have a cheap monotonic clock, so no caching thread is needed. Also add some relative time helpers | Dirk Engling |
| 5 days | While we're at it with the dynamic analyzers, add a tsan | Dirk Engling |
| 6 days | Add test for the revived keep alive | Dirk Engling |
| 6 days | Fix some tests to match retired stats modes and wrong config file ordering | Dirk Engling |
| 6 days | Add test for the correct order of partial scrapes | Dirk Engling |
| 6 days | Add test for the error packet when a hash is blacklisted (the part I forgot in last commit) | Dirk Engling |
| 6 days | Add test for the error packet when a hash is blacklisted | Dirk Engling |
