cimple - My little CI system (hopefully)

	Commit message (Collapse)	Author	Age
...
*	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...
*	ci_queue: rename a couple of functions	Egor Tensin	2023-05-12
\|
*	cmake: don't install() valgrind.sh	Egor Tensin	2023-05-07
\| \| \| \|	Turns out, I don't really need to install it for the tests.
*	valgrind: quiet mode	Egor Tensin	2023-05-07
\|
*	add tests to run binaries under Valgrind	Egor Tensin	2023-05-07
\| \| \| \| \| \| \| \| \| \| \|	This was quite a bit of refactoring in test/; everything should be more maintainable and robust in theory. Also, valgrind.sh was fixed to use exec (so that signals are passed to the underlying process); Valgrind command line options have also been tweaked. ./ci.sh fails now, but that should be fixable.
*	add a cimple-valgrind executable	Egor Tensin	2023-05-06
\|
*	add a TODO note	Egor Tensin	2023-05-06
\|
*	shut down server/workers gracefully on SIGTERM	Egor Tensin	2023-05-06
\|
*	get rid of __attribute__((constructor))	Egor Tensin	2023-05-06
\| \| \| \|	Explicit is better than implicit.
*	add a default SQLite database path	Egor Tensin	2023-04-29
\|
*	cmd_line: read executable name from /proc/self/exe	Egor Tensin	2023-04-29
\|
*	dedupe command line routines	Egor Tensin	2023-04-29
\|
*	make struct storage_settings_sqlite opaque	Egor Tensin	2023-04-29
\|
*	make struct ci_queue_entry opaque	Egor Tensin	2023-04-29
\|
*	net: use size instead of length for variable names	Egor Tensin	2023-04-29
\|
*	net: add struct buf	Egor Tensin	2023-04-29
\|
*	make struct client opaque	Egor Tensin	2023-04-29
\|
*	make struct worker opaque	Egor Tensin	2023-04-29
\|
*	make struct server opaque	Egor Tensin	2023-04-29
\|
*	make struct tcp_server opaque	Egor Tensin	2023-04-29
\|
*	fix a typo	Egor Tensin	2023-04-27
\|
*	rename commands	Egor Tensin	2023-04-27
\|
*	sqlite: fix error handling	Egor Tensin	2023-04-11
\|
*	put SQL headers to a separate directory	Egor Tensin	2023-04-11
\|
*	sqlite: create indices	Egor Tensin	2023-04-11
\|
*	actually create some tables in SQLite database	Egor Tensin	2023-04-11
\|
*	add copyright notices	Egor Tensin	2022-12-02
\|
*	create SQLite database on startup	Egor Tensin	2022-09-11
\|
*	log: refactoring	Egor Tensin	2022-09-08
\|
*	log: prepend timestamps	Egor Tensin	2022-09-08
\|
*	sanitize #include-s	Egor Tensin	2022-09-08
\|
*	ci: set some environment variables	Egor Tensin	2022-09-07
\|
*	net: removed unused API	Egor Tensin	2022-08-30
\|
*	net: error out unless net_recv_all receives all bytes	Egor Tensin	2022-08-30
\|
*	net: more portable	Egor Tensin	2022-08-30
\| \| \| \|	Use a predefined byte order, integers with fixed width, etc.
*	net: bind to IPv6	Egor Tensin	2022-08-30
\| \| \| \| \|	This makes sure we use the dual-stack feature to support both IPv4 and IPv6.
*	update command names	Egor Tensin	2022-08-28
\|
*	docker: shorten command values	Egor Tensin	2022-08-28
\|
*	server: notify workers about requeued jobs	Egor Tensin	2022-08-28
\| \| \| \|	This allows free workers to pick up jobs after dead workers.
*	worker: fix a crash	Egor Tensin	2022-08-28
\| \| \| \|	Found when running in Docker.
*	server: notify all threads about shutting down	Egor Tensin	2022-08-28
\| \| \| \| \| \| \|	The problem is pthread_cond_destroy is unsafe to call if there're threads waiting in pthread_cond_wait. I'm not sure this fix is enough: what if the "broadcast" doesn't reach the threads until we call pthread_cond_destroy? Does it even work that way? Idk
*	make proper "error" messages	Egor Tensin	2022-08-28
\| \| \| \| \|	Previously, the client had no way to distinguish errors from succesful calls.
*	fix Alpine builds	Egor Tensin	2022-08-28
\|
*	don't use the latest glibc features	Egor Tensin	2022-08-28
\| \| \| \| \|	pthread_attr_setsigmask_np is only available since 2.32, which is too modern.
*	add missing #include	Egor Tensin	2022-08-28
\|
*	cmake: add install() rules	Egor Tensin	2022-08-28
\|
*	make compilers happier	Egor Tensin	2022-08-28
\|
*	server: more logging	Egor Tensin	2022-08-28
\|
*	msg: fix a buffer overflow	Egor Tensin	2022-08-28
\| \| \| \|	Thanks, Valgrind!
*	holy crap, it actually kinda works now	Egor Tensin	2022-08-28
\| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \| \|	Previously, I had a stupid system where I would create a thread after every accept(), and put worker descriptors in a queue. A special "scheduler" thread would then pick them out, and give out jobs to complete. The problem was, of course, I couldn't conveniently poll job status from workers. I thought about using poll(), but that turned out to be a horribly complicated API. How do I deal with partial reads, for example? I don't honestly know. Then it hit me that I could just use the threads that handle accept()ed connections as "worker threads", which would synchronously schedule jobs and wait for them to complete. This solves every problem and removes the need for a lot of inter-thread synchronization magic. It even works now, holy crap! You can launch and terminate workers at will, and they will pick up new jobs automatically. As a side not, msg_recv_and_handle turned out to be too limiting and complicated for me, so I got rid of that, and do normal msg_recv/msg_send calls.