summaryrefslogtreecommitdiff
path: root/lib/libpthread/pthread.c
AgeCommit message (Expand)Author
2007-10-16... but preserve the linked list, for the debugger only.ad
2007-10-16Replace the global thread list with a red-black tree. From joerg@.ad
2007-09-11Put new threads on the tail of pthread__allqueue, for the debugger.ad
2007-09-08- Get rid of self->pt_mutexhint and use pthread__mutex_owned() instead.ad
2007-09-07- Don't take the mutex's spinlock (ptr_interlock) in pthread_cond_wait().ad
2007-08-23Acquire the correct lock in pthread_detach(). Spotted by Jan Kryl.ad
2007-08-17pthread_suspend_np, pthread_resume_np, pthread_detach: return correct codead
2007-08-16Trim fat off libpthread internal spinlock operations. Makes a mesurablead
2007-08-16- Reinitialize the absolute minimum when recycling user thread state.ad
2007-08-16Remove PT_FIXEDSTACKSIZE_LG.ad
2007-08-16Cache thread context for creation instead of setting it up every time.ad
2007-08-15Simplify pthread_join(). Knocks about 5% off pthread_create/join in a loop.ad
2007-08-07Missed variable rename..ad
2007-08-07Change the signature of _lwp_park() to accept an lwpid_t and secondad
2007-08-04Make libpthread_dbg build again.ad
2007-08-04Fix error in previous.ad
2007-08-04Some significant performance improvements, and a fix for a race with pthreadad
2007-03-24- Test+branch is usually cheaper than making an indirect function call,ad
2007-03-14- pthread__park: always check for cancellation.ad
2007-03-05- Simplify the interface to pthread__park() and friends slightly.ad
2007-03-02Silence a gcc/lint warning.ad
2007-03-02Remove the PTHREAD_SA option. If M:N threads is reimplemented it'sad
2007-03-02pthread_detach: acquire the join lock before changing pt_flags.ad
2007-02-21- Put detached & exiting threads on the tail of the deadqueue, not head.ad
2007-02-21Minor optimisation to previous: only check for continued existence ofad
2007-02-15fix a race between pthread_exit and pthread_create.yamt
2007-02-09Cast the return value of _lwp_unpark_all(), just in case.ad
2007-01-31use __func__ instead of __FUNCTION__ which is a gcc extension.christos
2007-01-20pthread_create(): clear newthread->pt_sleeponq before calling _lwp_create().ad
2007-01-20add a missing volatile.christos
2007-01-16Fix a race between pthread_exit() and pthread_join().ad
2007-01-16Fix a race between pthread_create() and pthread_exit() in the 1:1 case.ad
2007-01-16Pass in the 'hint' argument when parking/unparking LWPs.ad
2006-12-24Fix bugs with and improve upon previous.ad
2006-12-23Add another comment.ad
2006-12-23Add a comment.ad
2006-12-23Conditionalised support for 1:1 threads. Needs associated kernel changesad
2006-04-24There is no point in using the internal sigprocmask() to initializedrochner
2006-02-12- do PTHREAD_MLOCK_KLUDGE in pthread__stackid_setup, rather than callers,yamt
2005-10-19if mlock() fails in pthread_create(), return EAGAIN instead ofchs
2005-10-19starting the pthread library (ie. calling pthread__start()) beforechs
2005-10-16in pthread_kill() and pthread_suspend_np(), return without doing anythingchs
2005-10-16in pthread_mutex_lock_slow(), pthread_rwlock_timedrdlock() and sem_wait(),chs
2005-07-01make this compile without PTHREAD_MLOCK_KLUDGE.yamt
2005-02-26Keep the kernel updated with signal action signal masks (act.sa_mask) untilnathanw
2005-02-10Default back pthread__concurrency to 1 until we fix the regression testchristos
2005-02-10If $PTHREAD_CONCURRENCY is not set, set it to ncpuchristos
2005-02-03Add charles' mlock fixes, protected with PTHREAD_MLOCK_KLUDGE and enabledchristos
2005-01-06Add some additional assertions and debugging printf()s.mycroft
2004-08-12- fix pthread_detach with an already exit'ed thread.yamt