| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2005-12-19 | sanity check for key_delete was missing ! | Ted Unangst | |
| 2005-12-18 | initialize all spinlocks to _SPINLOCK_UNLOCKED | Ted Unangst | |
| 2005-12-17 | Document PT_STEP (better late than never), and group the machine-dependent | Miod Vallat | |
| request with a (fixed) list of arches for each; tweaks and ok jmc@ kettenis@ | |||
| 2005-12-17 | sure it is not perfect, but at least we can compile rthreads on hppa. | Marco Peereboom | |
| ok deraadt@ | |||
| 2005-12-16 | libresolv.a is no longer needed | Theo de Raadt | |
| 2005-12-16 | boolshit | Michael Shalayeff | |
| 2005-12-16 | clarify sentence; cannot reuse a stream socket after even trying to do a ↵ | Theo de Raadt | |
| connect; spotted by tedu | |||
| 2005-12-14 | My email address changed quite a while ago... | Dale Rahn | |
| 2005-12-14 | Initial versions of arm code, in-tree development (not yet working). | Dale Rahn | |
| 2005-12-14 | Add locking primitive for hppa. ok kettenis@ | Marco Peereboom | |
| 2005-12-14 | all is made clear: pthread_cond_timedwait takes absolute timeouts | Ted Unangst | |
| 2005-12-14 | add pthread_cleanup_push and pthread_cleanup_pop | Ted Unangst | |
| 2005-12-14 | add pthread_kill (just kill(2) really) | Ted Unangst | |
| 2005-12-14 | the beatings will continue until morale improves | Ted Unangst | |
| 2005-12-14 | there should be a lock around the tls key table | Ted Unangst | |
| 2005-12-14 | add bits for pthread_cancel. we don't really have cancellation points yet | Ted Unangst | |
| but some of the functions are here now. | |||
| 2005-12-14 | check for waiters when destroying a mutex or semaphore | Ted Unangst | |
| 2005-12-14 | change keys to use table instead of list, makes a sane destructor implementation | Ted Unangst | |
| possible | |||
| 2005-12-13 | make the mutex unlock and sleep in pthread_cond_wait properly atomic | Ted Unangst | |
| 2005-12-13 | several silly bugs in pthread_cond_timedwait. | Ted Unangst | |
| 1. the time was off by a factor of 10 2. wouldn't return error code if timeout was reached 3. the big one. thrsleep syscall doesn't return EWOULDBLOCK. it returns -1 and puts the error in errno. doh. | |||
| 2005-12-13 | update thrsleep and thrwakeup - first arg changed from long to void * | Ted Unangst | |
| 2005-12-13 | correct implementation of pthread_cond_signal. it doesn't raise the sem | Ted Unangst | |
| value if there are no waiters. | |||
| 2005-12-13 | Remove the advertising clause in the UCB license which Berkeley | Jonathan Gray | |
| rescinded 22 July 1999. Checked by ian@ and deraadt@ | |||
| 2005-12-10 | because lint says so; ok espie | Theo de Raadt | |
| 2005-12-07 | add the posix semaphore functions. this lets vlc work. | Ted Unangst | |
| ok brad | |||
| 2005-12-07 | malloc the right size, and memset after malloc | Ted Unangst | |
| 2005-12-06 | add pthread_once. unfortunately, the public pthread.h header | Ted Unangst | |
| defines the pthread_once_t internals, so we're stuck with them. | |||
| 2005-12-06 | can't build shared lib without shlib_version | Ted Unangst | |
| Maxim Bourmistrov noticed. | |||
| 2005-12-04 | copy alpha atomic_lock implementation from libpthread. | Brad Smith | |
| 2005-12-04 | add a powerpc implementation of rfork_thread(). | Brad Smith | |
| From Tim Wiess <tim at nop dot cx> ok drahn@ | |||
| 2005-12-04 | copy powerpc atomic_lock implementation from libpthread. | Brad Smith | |
| ok drahn@ | |||
| 2005-12-03 | rcs for the makefile | Ted Unangst | |
| 2005-12-03 | syscall is actually sched_yield now, as millert suggested | Ted Unangst | |
| 2005-12-03 | add userland thread library. incomplete, but functional | Ted Unangst | |
| 2005-12-02 | make write-end of signal socketpair non-blocking. | Theo de Raadt | |
| every signal received writes 1 byte to the socketpair. if you are outside the libevent main loop too long, the socketpair might fill up. that write would then block, in a signal handler. related to this, yesterdy we changed the signal handler to not trash errno. as for the read end, there will be multiple libevent reads off the socketpair (100 bytes at a time) until the socketpair is empty again ok provos | |||
| 2005-12-02 | pollops should be const | Theo de Raadt | |
| 2005-12-02 | lookup out of evsigcaught[] must be sig_atomic_t too | Theo de Raadt | |
| 2005-12-02 | never say unsigned; say at least unsigned int | Theo de Raadt | |
| 2005-12-01 | save errno in signal handler since it does a system call | Theo de Raadt | |
| only access sig_atomic_t variables, to make it further save ok aaron | |||
| 2005-11-30 | know __arm__ too | Theo de Raadt | |
| 2005-11-30 | timezone -> time zone | Jason McIntyre | |
| 2005-11-30 | Use sysctl(KERN_ARND) to get n bytes, instead of just 4 at a time | Otto Moerbeek | |
| and remove fallback code. If somebody is dumb enough to make the sysctl fail using systrace, he deserves what he gets. Saves 7 syscalls on process startup. looks good miod@ ok deraadt@ tedu@ | |||
| 2005-11-29 | if we pull in a .S file, we must fake out the lint with a .c file | Theo de Raadt | |
| for this first cut, we will do this for alloca() using alloca.c by adding it to LSRCS | |||
| 2005-11-29 | Provide stubs for things which are supplied by crt0. I don't think we | Theo de Raadt | |
| want a crt0.ln, at least, not yet. We may reconsider later. | |||
| 2005-11-29 | document KERN_ARND better; ok deraadt@ | Otto Moerbeek | |
| 2005-11-28 | spaces, and a off_t typecast to please lint | Theo de Raadt | |
| 2005-11-28 | some nice off_t casts to please lint | Theo de Raadt | |
| 2005-11-28 | unused arg in internal static API | Theo de Raadt | |
| 2005-11-28 | ARGSUSED stubs | Theo de Raadt | |
| 2005-11-28 | make Lint prototypes equal; ok millert | Theo de Raadt | |
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |