Commit message (Collapse) | Author | Age | ||
---|---|---|---|---|
... | ||||
* | valgrind: exit with an error after leak detection | Egor Tensin | 2023-06-28 | |
| | | | | I cannot believe this isn't default. | |||
* | file: readlink(2) doesn't null-terminate | Egor Tensin | 2023-06-28 | |
| | | | | Thanks, valgrind. | |||
* | minor refactoring | Egor Tensin | 2023-06-14 | |
| | ||||
* | cmake: fix grammar in option description | Egor Tensin | 2023-06-13 | |
| | ||||
* | cmake: why the hell would I need MSVC options | Egor Tensin | 2023-06-13 | |
| | ||||
* | minor refactoring | Egor Tensin | 2023-06-13 | |
| | ||||
* | signal: remove the stupid add_to_event_loop wrapper | Egor Tensin | 2023-06-13 | |
| | ||||
* | server: handle disconnected workers gracefully | Egor Tensin | 2023-06-13 | |
| | ||||
* | event_loop: support more event types | Egor Tensin | 2023-06-13 | |
| | ||||
* | event_loop: add event_loop_add_once | Egor Tensin | 2023-06-13 | |
| | ||||
* | signal: sigprocmask -> pthread_sigmask | Egor Tensin | 2023-06-13 | |
| | ||||
* | signal: refactoring | Egor Tensin | 2023-06-13 | |
| | ||||
* | sanitize #include-s | Egor Tensin | 2023-06-13 | |
| | ||||
* | signal: remove unused APIs | Egor Tensin | 2023-06-13 | |
| | ||||
* | event_loop: hide the API details | Egor Tensin | 2023-06-13 | |
| | ||||
* | use signalfd to stop on SIGTERM | Egor Tensin | 2023-06-13 | |
| | | | | | | | | | | | | | | | | | | Is this an overkill? I don't know. The thing is, correctly intercepting SIGTERM (also SIGINT, etc.) is incredibly tricky. For example, before this commit, my I/O loops in server.c and worker.c were inherently racy. This was immediately obvious if you tried to run the tests. The tests (especially the Valgrind flavour) would run a worker, wait until it prints a "Waiting for a new command" line, and try to kill it using SIGTERM. The problem is, the global_stop_flag check could have already been executed by the worker, and it would hang forever in recv(). The solution seems to be to use signalfd and select()/poll(). I've never used either before, but it seems to work well enough - at least the very same tests pass and don't hang now. | |||
* | log: refactoring | Egor Tensin | 2023-06-12 | |
| | ||||
* | log: thread-safe logging | Egor Tensin | 2023-06-12 | |
| | ||||
* | log: prefix with thread ID | Egor Tensin | 2023-06-12 | |
| | ||||
* | command: fix a race condition | Egor Tensin | 2023-06-12 | |
| | | | | | Modifying cmd_dispatcher fields like that make it inherently unsafe to call cmd_dispatcher_handle_conn concurrently. | |||
* | sqlite: fix generate-sql-header.py | Egor Tensin | 2023-06-12 | |
| | ||||
* | msg: rework some APIs | Egor Tensin | 2023-06-11 | |
| | ||||
* | worker_queue: fix worker_queue_add_last | Egor Tensin | 2023-06-09 | |
| | ||||
* | STAILQ -> SIMPLEQ | Egor Tensin | 2023-06-09 | |
| | ||||
* | msg: add msg_send_from_argv shortcut | Egor Tensin | 2023-06-09 | |
| | ||||
* | add {file,net}_close as a wrapper to close(2) | Egor Tensin | 2023-06-09 | |
| | ||||
* | fix realloc usage | Egor Tensin | 2023-05-29 | |
| | | | | | | | | Remember, this is always a mistake: ptr = realloc(ptr, size); You still need to free() the original ptr if realloc fails. | |||
* | file: rework file_read | Egor Tensin | 2023-05-17 | |
| | | | | | It now increases the buffer size exponentially until it finishes reading the file. | |||
* | signal: refactoring | Egor Tensin | 2023-05-15 | |
| | ||||
* | net: drop the _all suffix | Egor Tensin | 2023-05-15 | |
| | ||||
* | msg: strings, not words | Egor Tensin | 2023-05-15 | |
| | ||||
* | minor refactoring | Egor Tensin | 2023-05-15 | |
| | ||||
* | signal: refactoring, add comments in tcp_server, etc. | Egor Tensin | 2023-05-15 | |
| | ||||
* | valgrind: track open file descriptors | Egor Tensin | 2023-05-15 | |
| | ||||
* | EINVAL means EINTR also? | Egor Tensin | 2023-05-15 | |
| | ||||
* | worker_queue: forgot to close file descriptors in worker_destroy | Egor Tensin | 2023-05-15 | |
| | ||||
* | cmake: disable -Werror for debug builds | Egor Tensin | 2023-05-15 | |
| | ||||
* | rework server-worker communication | Egor Tensin | 2023-05-15 | |
| | | | | | | | | | | | | | | | | | | OK, this is a major rework. * tcp_server: connection threads are not detached anymore, the caller has to clean them up. This was done so that the server can clean up the threads cleanly. * run_queue: simple refactoring, run_queue_entry is called just run now. * server: worker threads are now killed when a run is assigned to a worker. * worker: the connection to server is no longer persistent. A worker sends "new-worker", waits for a task, closes the connection, and when it's done, sends the "complete" message and waits for a new task. This is supposed to improve resilience, since the worker-server connections don't have to be maintained while the worker is doing a CI run. | |||
* | command: adjust order of parameters to handlers | Egor Tensin | 2023-05-14 | |
| | ||||
* | process: add process_output_dump | Egor Tensin | 2023-05-14 | |
| | ||||
* | msg: add functions for one-off communication | Egor Tensin | 2023-05-14 | |
| | ||||
* | msg: enforce at least one word | Egor Tensin | 2023-05-14 | |
| | | | | Also, move some stuff to net.c where it belongs. | |||
* | fix unnecessary #include-s | Egor Tensin | 2023-05-13 | |
| | ||||
* | ci_queue -> run_queue | Egor Tensin | 2023-05-13 | |
| | | | | Also, some minor refactoring. | |||
* | cmake: document _GNU_SOURCE usage | Egor Tensin | 2023-05-13 | |
| | ||||
* | use -std=c17 -Wpedantic | Egor Tensin | 2023-05-13 | |
| | ||||
* | command: refactoring | Egor Tensin | 2023-05-13 | |
| | ||||
* | best practices & coding style fixes | Egor Tensin | 2023-05-13 | |
| | | | | | | | | * I don't really need to declare all variables at the top of the function anymore. * Default-initialize variables more. * Don't set the output parameter until the object is completely constructed. | |||
* | add command module to handle request-response communications | Egor Tensin | 2023-05-13 | |
| | ||||
* | ci_queue: fix a broken getter | Egor Tensin | 2023-05-13 | |
| | | | | | | Everything was broken starting from the "making struct ci_queue_entry opaque" commit. Damn, I really wish I'd have some kind of automated testing to catch errors like this... |