summary refs log tree commit diff
path: root/rt/aio.h
diff options
context:
space:
mode:
authorUlrich Drepper <drepper@redhat.com>2000-01-05 02:09:12 +0000
committerUlrich Drepper <drepper@redhat.com>2000-01-05 02:09:12 +0000
commit1d2fc9b3c59d0e83e04139ddf633731264b76ea2 (patch)
treec738cf2a40851dc25be2c252ba5dbb7f335b5e14 /rt/aio.h
parentf19f2b34439145daf300bf12789bbc61c8d4db28 (diff)
downloadglibc-1d2fc9b3c59d0e83e04139ddf633731264b76ea2.tar.gz
glibc-1d2fc9b3c59d0e83e04139ddf633731264b76ea2.tar.xz
glibc-1d2fc9b3c59d0e83e04139ddf633731264b76ea2.zip
Redesigned how cancellation unblocks a thread from internal cancellation points (sem_wait, pthread_join, pthread_cond_{wait,timedwait}). Cancellation won't eat a signal in any of these functions (*required* by POSIX and Single Unix Spec!).
2000-01-03  Kaz Kylheku  <kaz@ashi.footprints.net>

	Redesigned how cancellation unblocks a thread from internal
	cancellation points (sem_wait, pthread_join,
	pthread_cond_{wait,timedwait}).
	Cancellation won't eat a signal in any of these functions
	(*required* by POSIX and Single Unix Spec!).
	* condvar.c: spontaneous wakeup on pthread_cond_timedwait won't eat a
	simultaneous condition variable signal (not required by POSIX
	or Single Unix Spec, but nice).
	* spinlock.c: __pthread_lock queues back any received restarts
	that don't belong to it instead of assuming ownership of lock
	upon any restart; fastlock can no longer be acquired by two threads
	simultaneously.
	* restart.h: restarts queue even on kernels that don't have
	queued real time signals (2.0, early 2.1), thanks to atomic counter,
	avoiding a rare race condition in pthread_cond_timedwait.
Diffstat (limited to 'rt/aio.h')
0 files changed, 0 insertions, 0 deletions