like mutexes and semaphores, rwlocks suffered from a race condition
where the unlock operation could access the lock memory after another
thread successfully obtained the lock (and possibly destroyed or
unmapped the object). this has been fixed in the same way it was fixed
for other lock types.
in addition, the previous implementation favored writers over readers.
in the absence of other considerations, that is the best behavior for
rwlocks, and posix explicitly allows it. however posix also requires
read locks to be recursive. if writers are favored, any attempt to
obtain a read lock while a writer is waiting for the lock will fail,
causing "recursive" read locks to deadlock. this can be avoided by
keeping track of which threads already hold read locks, but doing so
requires unbounded memory usage, and there must be a fallback case
that favors readers in case memory allocation failed. and all of this
must be synchronized. the cost, complexity, and risk of errors in
getting it right is too great, so we simply favor readers.
tracking of the owner of write locks has been removed, as it was not
useful for anything. it could allow deadlock detection, but it's not
clear to me that returning EDEADLK (which a buggy program is likely to
ignore) is better than deadlocking; at least the latter behavior
prevents further data corruption. a correct program cannot invoke this
situation anyway.
the reader count and write lock state, as well as the "last minute"
waiter flag have all been combined into a single atomic lock. this
means all state transitions for the lock are atomic compare-and-swap
operations. this makes establishing correctness much easier and may
improve performance.
finally, some code duplication has been cleaned up. more is called
for, especially the standard __timedwait idiom repeated in all locks.
#define _m_count __u.__i[5]
#define _c_block __u.__i[0]
#define _c_clock __u.__i[1]
#define _m_count __u.__i[5]
#define _c_block __u.__i[0]
#define _c_clock __u.__i[1]
-#define _rw_wrlock __u.__i[0]
-#define _rw_readers __u.__i[1]
-#define _rw_waiters __u.__i[2]
-#define _rw_owner __u.__i[3]
+#define _rw_lock __u.__i[0]
+#define _rw_waiters __u.__i[1]
#define _b_inst __u.__p[0]
#define _b_limit __u.__i[2]
#define _b_lock __u.__i[3]
#define _b_inst __u.__p[0]
#define _b_limit __u.__i[2]
#define _b_lock __u.__i[3]
int pthread_rwlock_rdlock(pthread_rwlock_t *rw)
{
int pthread_rwlock_rdlock(pthread_rwlock_t *rw)
{
- while (pthread_rwlock_tryrdlock(rw))
- __wait(&rw->_rw_wrlock, &rw->_rw_waiters, 1, 0);
- return 0;
+ return pthread_rwlock_timedrdlock(rw, 0);
int pthread_rwlock_timedrdlock(pthread_rwlock_t *rw, const struct timespec *at)
{
int pthread_rwlock_timedrdlock(pthread_rwlock_t *rw, const struct timespec *at)
{
- int w=0;
- while (pthread_rwlock_tryrdlock(rw)) {
- if (!w) a_inc(&rw->_rw_waiters), w++;
- if (__timedwait(&rw->_rw_wrlock, 1, CLOCK_REALTIME, at, 0, 0, 0)==ETIMEDOUT) {
- if (w) a_dec(&rw->_rw_waiters);
- return ETIMEDOUT;
- }
+ int r, t;
+ while ((r=pthread_rwlock_tryrdlock(rw))==EBUSY) {
+ if (!(r=rw->_rw_lock) || (r&0x7fffffff)!=0x7fffffff) continue;
+ t = r | 0x80000000;
+ a_inc(&rw->_rw_waiters);
+ a_cas(&rw->_rw_lock, r, t);
+ r = __timedwait(&rw->_rw_lock, t, CLOCK_REALTIME, at, 0, 0, 0);
+ a_dec(&rw->_rw_waiters);
+ if (r && r != EINTR) return r;
- if (w) a_dec(&rw->_rw_waiters);
- return 0;
int pthread_rwlock_timedwrlock(pthread_rwlock_t *rw, const struct timespec *at)
{
int pthread_rwlock_timedwrlock(pthread_rwlock_t *rw, const struct timespec *at)
{
- int nr, *p, w=0;
- while (pthread_rwlock_trywrlock(rw)==EAGAIN) {
- if (!w) a_inc(&rw->_rw_waiters), w++;
- if ((nr=rw->_rw_readers)) p = &rw->_rw_readers;
- else nr=1, p = &rw->_rw_wrlock;
- if (__timedwait(p, nr, CLOCK_REALTIME, at, 0, 0, 0)==ETIMEDOUT) {
- if (w) a_dec(&rw->_rw_waiters);
- return ETIMEDOUT;
- }
+ int r, t;
+ while ((r=pthread_rwlock_trywrlock(rw))==EBUSY) {
+ if (!(r=rw->_rw_lock)) continue;
+ t = r | 0x80000000;
+ a_inc(&rw->_rw_waiters);
+ a_cas(&rw->_rw_lock, r, t);
+ r = __timedwait(&rw->_rw_lock, t, CLOCK_REALTIME, at, 0, 0, 0);
+ a_dec(&rw->_rw_waiters);
+ if (r && r != EINTR) return r;
- if (w) a_dec(&rw->_rw_waiters);
- return 0;
int pthread_rwlock_tryrdlock(pthread_rwlock_t *rw)
{
int pthread_rwlock_tryrdlock(pthread_rwlock_t *rw)
{
- a_inc(&rw->_rw_readers);
- if (rw->_rw_wrlock) {
- a_dec(&rw->_rw_readers);
- if (rw->_rw_waiters && !rw->_rw_readers)
- __wake(&rw->_rw_readers, 1, 0);
- return EBUSY;
- }
+ int val, cnt;
+ do {
+ val = rw->_rw_lock;
+ cnt = val & 0x7fffffff;
+ if (cnt == 0x7fffffff) return EBUSY;
+ if (cnt == 0x7ffffffe) return EAGAIN;
+ } while (a_cas(&rw->_rw_lock, val, val+1) != val);
int pthread_rwlock_trywrlock(pthread_rwlock_t *rw)
{
int pthread_rwlock_trywrlock(pthread_rwlock_t *rw)
{
- if (a_xchg(&rw->_rw_wrlock, 1))
- return EBUSY;
- if (rw->_rw_readers) {
- a_store(&rw->_rw_wrlock, 0);
- return EBUSY;
- }
- rw->_rw_owner = pthread_self()->tid;
+ if (a_cas(&rw->_rw_lock, 0, 0x7fffffff)) return EBUSY;
int pthread_rwlock_unlock(pthread_rwlock_t *rw)
{
int pthread_rwlock_unlock(pthread_rwlock_t *rw)
{
- struct pthread *self = pthread_self();
- if (rw->_rw_owner == self->tid) {
- rw->_rw_owner = 0;
- a_store(&rw->_rw_wrlock, 0);
- if (rw->_rw_waiters)
- __wake(&rw->_rw_wrlock, -1, 0);
- return 0;
- }
- a_dec(&rw->_rw_readers);
- if (rw->_rw_waiters && !rw->_rw_readers)
- __wake(&rw->_rw_readers, 1, 0);
+ int val, cnt, waiters, new;
+
+ do {
+ val = rw->_rw_lock;
+ cnt = val & 0x7fffffff;
+ waiters = rw->_rw_waiters;
+ new = (cnt == 0x7fffffff || cnt == 1) ? 0 : val-1;
+ } while (a_cas(&rw->_rw_lock, val, new) != val);
+
+ if (!new && (waiters || val<0))
+ __wake(&rw->_rw_lock, 1, 0);
+
int pthread_rwlock_wrlock(pthread_rwlock_t *rw)
{
int pthread_rwlock_wrlock(pthread_rwlock_t *rw)
{
- int nr;
- while (pthread_rwlock_trywrlock(rw)==EAGAIN) {
- if ((nr=rw->_rw_readers))
- __wait(&rw->_rw_readers, &rw->_rw_waiters, nr, 0);
- else
- __wait(&rw->_rw_wrlock, &rw->_rw_waiters, 1, 0);
- }
- return 0;
+ return pthread_rwlock_timedwrlock(rw, 0);
projects / musl / commitdiff