src - OpenBSD base system

Age	Commit message (Collapse)	Author
2012-09-08	Link init_priority into the regress tree now that it passes (except	Matthew Dempsky
	mysteriously on m88k).
2012-09-07	Give this a chance to build with gcc2. ok matthew@	Miod Vallat

2012-09-06	Add a (currently failing) regress test for GNU C++'s init_priority	Matthew Dempsky
	attribute. See "info --index-search=init_priority gcc" for details on how this feature is supposed to work.
2012-09-06	Move the commented out test program in pkcs5_pbkdf2.c into a proper	Matthew Dempsky
	regress test.
2012-09-05	run regress with the local env too	Eric Faurot

2012-09-05	use LDSTATIC for building static executables	Eric Faurot

2012-08-22	correct check.	Federico G. Schwindt

2012-08-19	document some more disabled tests.	Federico G. Schwindt

2012-08-19	Switch to -lpthread.	Federico G. Schwindt

2012-08-07	allow to choose an alternate regress script	Eric Faurot

2012-08-04	Add a test for caching of stacks with the default attributes	Philip Guenthe

2012-07-29	add a flag to choose between gethostbyname and gethostbyaddr	Eric Faurot

2012-07-13	import regression suite for asr	Eric Faurot

2012-07-11	catch off-by-one errors in stpncpy(); ok guenther@	Christian Weisgerber

2012-07-11	Now that socket timeouts don't return early, fix the regress to match	Philip Guenthe

2012-07-08	Enhance the blocking test and add a test for closing while half-open	Philip Guenthe

2012-05-03	Add pthread spinlock support.	Paul Irofti
	Implementation, documentation and naive regression tests for: - pthread_spin_init() - pthread_spin_destroy() - pthread_spin_lock() - pthread_spin_trylock() - pthread_spin_unlock() Implementation okay guenther@, documentation okay jmc@.
2012-04-13	Remove suspend_np1 test from regress	Kurt Miller

2012-04-13	Remove suspend_np1 from regress build.	Kurt Miller

2012-04-13	Make it easier to compile regress tests on other OSes	Philip Guenthe

2012-04-11	Add a simple barrier regression test.	Paul Irofti

2012-03-24	Another tiny test case	Philip Guenthe

2012-03-19	Add blocked_fifo.	Owain Ainsworth

2012-03-19	Add a (failing) test for a deadlock with fifos and threads.	Owain Ainsworth
	The ``sleep until we have a writer'' behaviour of an open() on a fifo does so with the file descriptor table locked, so if we are waiting for another thread to be our writer we will hang forever. Found this using zotero and firefox. ok fgsch@
2012-03-07	another questionable hang.	Federico G. Schwindt

2012-03-03	Add tests for sem_timedwait()	Philip Guenthe

2012-03-03	Verify that sem_wait() doesn't return zero without decrementing the	Philip Guenthe
	semaphore when a signal is received. It should either resume waiting or return EINTR, preferrably the former.
2012-03-03	While we guarantee that a zero-filled sem_t variable will trigger	Philip Guenthe
	EINVAL errors, a sem_t on the stack may have garbage, so make the tests portable by moving the sem_t from function to file scope. problem found by miod@
2012-02-27	test based on python's dubious test_3_join_in_forked_from_thread test.	Federico G. Schwindt

2012-02-26	update error message. inetd has not been used since rev 1.3.	Federico G. Schwindt
	spotted by miod@
2012-02-26	similar change to the setsockopt1.c one: use the resolution of the monotonic	Federico G. Schwindt
	clock from clock_getres(). while here use a different port when binding.
2012-02-26	Let this compile with gcc 2.	Miod Vallat

2012-02-23	Check that PTHREAD_MUTEX_NORMAL mutexes will self-deadlock and that	Philip Guenthe
	pthread_mutex_timedlock() will block and then timeout on self-deadlock
2012-02-22	similar change to the setsockopt1.c one: use the resolution of the monotonic	Federico G. Schwindt
	clock from clock_getres(). while here use a different port when binding.
2012-02-21	Add pthread_suspend_np regress test. okay guenther@	Kurt Miller

2012-02-20	Switch the time measuring to be in timespecs, and then check the	Philip Guenthe
	difference against the resolution of the monotonic clock from clock_getres() discussed with fgsch@
2012-02-20	- remove invalid potions of blocked_* regress tests	Kurt Miller
	- reenable blocked_* regress tests. while this behaivor is not specifited by posix, deadlocking is not a reasonable behaivor and some ports depend on this to work.
2012-02-20	Disable the dup2_race test, as it requires a buggy behavior that uthreads	Philip Guenthe
	happened to exhibit
2012-02-20	Round-robin behavior isn't guaranteed by SCHED_OTHER and the kernel	Philip Guenthe
	doesn't have SCHED_FIFO or SCHED_RR support, so use sleep(1) to 'ensure' the second thread gets the lock when expected
2012-02-20	Enable the errno test; add a quick rwlocks test	Philip Guenthe

2012-02-20	End the output with a newline	Philip Guenthe

2012-02-20	This test needs a large nprocs ulimit, so just have it raise its own	Philip Guenthe

2012-02-20	Need to block the signals in all thread for the kill(getpid(), SIGFOO)	Philip Guenthe
	tests to work. Disable the "sigwait for blocked but ignored signal" test as it depended on undefined behavior (POSIX quoted included for reference) and rthreads (currently) handles it the other way.
2012-02-20	SIGSTOP, SIGKILL, and (for threaded programs) SIGTHR are unblockable	Philip Guenthe

2012-02-20	pthread_sigmask() was fixed to return the errno value on failure	Philip Guenthe

2012-02-20	ops. compare against the socket timeout, not the alarm one.	Federico G. Schwindt
	sometimes these tests are incorrectly marked as failing (ktrace shows the read is returning within the specified timeout) but there are other times where they're taking slightly longer than expected. for now both cases are marked as failure.
2012-02-19	fix tests.	Federico G. Schwindt

2012-02-19	Add Id and public-domain declaration	Philip Guenthe

2012-02-19	Add regress test for stack attributes	Philip Guenthe
	Disable some tests that depended on uthread's particular undefined behavior
2012-01-04	link new semaphore tests	Martin Pieuchot