4 files changed, 76 insertions, 10 deletions

diff --git a/src/thread/pthread_barrier_destroy.c b/src/thread/pthread_barrier_destroy.c
index 2898c41a..f82664d5 100644
--- a/src/thread/pthread_barrier_destroy.c
+++ b/src/thread/pthread_barrier_destroy.c
@@ -1,6 +1,12 @@
 #include "pthread_impl.h"
 
+void __vm_lock(int), __vm_unlock(void);
+
 int pthread_barrier_destroy(pthread_barrier_t *b)
 {
+	if (b->_b_limit < 0) {
+		__vm_lock(-1);
+		__vm_unlock();
+	}
 	return 0;
 }
diff --git a/src/thread/pthread_barrier_init.c b/src/thread/pthread_barrier_init.c
index 736d1014..ccaab4eb 100644
--- a/src/thread/pthread_barrier_init.c
+++ b/src/thread/pthread_barrier_init.c
@@ -2,7 +2,7 @@
 
 int pthread_barrier_init(pthread_barrier_t *b, const pthread_barrierattr_t *a, unsigned count)
 {
-	if (!count) return EINVAL;
-	*b = (pthread_barrier_t){ ._b_limit = count-1 };
+	if (count-1 > INT_MAX-1) return EINVAL;
+	*b = (pthread_barrier_t){ ._b_limit = count-1 | (a?*a:0) };
 	return 0;
 }
diff --git a/src/thread/pthread_barrier_wait.c b/src/thread/pthread_barrier_wait.c
index aed1adc8..db432ba3 100644
--- a/src/thread/pthread_barrier_wait.c
+++ b/src/thread/pthread_barrier_wait.c
@@ -1,5 +1,62 @@
 #include "pthread_impl.h"
 
+static int vmlock[2];
+
+void __vm_lock(int inc)
+{
+	for (;;) {
+		int v = vmlock[0];
+		if (inc*v < 0) __wait(vmlock, vmlock+1, v, 1);
+		else if (a_cas(vmlock, v, v+inc)==v) break;
+	}
+}
+
+void __vm_unlock(void)
+{
+	if (vmlock[0]>0) a_dec(vmlock);
+	else a_inc(vmlock);
+	if (vmlock[1]) __wake(vmlock, 1, 1);
+}
+
+static int pshared_barrier_wait(pthread_barrier_t *b)
+{
+	int limit = (b->_b_limit & INT_MAX) + 1;
+	int seq;
+	int ret = 0;
+
+	if (!limit) return PTHREAD_BARRIER_SERIAL_THREAD;
+
+	while (a_swap(&b->_b_lock, 1))
+		__wait(&b->_b_lock, &b->_b_waiters, 1, 0);
+
+	seq = b->_b_seq;
+
+	if (++b->_b_count == limit) {
+		ret = PTHREAD_BARRIER_SERIAL_THREAD;
+		b->_b_seq++;
+		__wake(&b->_b_seq, -1, 0);
+	} else {
+		a_store(&b->_b_lock, 0);
+		if (b->_b_waiters) __wake(&b->_b_lock, 1, 0);
+		__wait(&b->_b_seq, 0, seq, 0);
+	}
+
+	__vm_lock(+1);
+
+	if (a_fetch_add(&b->_b_count, -1)==1) {
+		b->_b_seq++;
+		__wake(&b->_b_seq, -1, 0);
+		a_store(&b->_b_lock, 0);
+		if (b->_b_waiters) __wake(&b->_b_lock, 1, 0);
+	} else {
+		__wait(&b->_b_seq, 0, seq+1, 0);
+	}
+	
+	__vm_unlock();
+
+	return 0;
+}
+
 struct instance
 {
 	int count;
@@ -16,9 +73,12 @@ int pthread_barrier_wait(pthread_barrier_t *b)
 	/* Trivial case: count was set at 1 */
 	if (!limit) return PTHREAD_BARRIER_SERIAL_THREAD;
 
+	/* Process-shared barriers require a separate, inefficient wait */
+	if (limit < 0) return pshared_barrier_wait(b);
+
 	/* Otherwise we need a lock on the barrier object */
 	while (a_swap(&b->_b_lock, 1))
-		__wait(&b->_b_lock, &b->_b_waiters, 1, 0);
+		__wait(&b->_b_lock, &b->_b_waiters, 1, 1);
 	inst = b->_b_inst;
 
 	/* First thread to enter the barrier becomes the "instance owner" */
@@ -27,7 +87,7 @@ int pthread_barrier_wait(pthread_barrier_t *b)
 		int spins = 10000;
 		b->_b_inst = inst = &new_inst;
 		a_store(&b->_b_lock, 0);
-		if (b->_b_waiters) __wake(&b->_b_lock, 1, 0);
+		if (b->_b_waiters) __wake(&b->_b_lock, 1, 1);
 		while (spins-- && !inst->finished)
 			a_spin();
 		a_inc(&inst->finished);
@@ -40,19 +100,19 @@ int pthread_barrier_wait(pthread_barrier_t *b)
 	if (++inst->count == limit) {
 		b->_b_inst = 0;
 		a_store(&b->_b_lock, 0);
-		if (b->_b_waiters) __wake(&b->_b_lock, 1, 0);
+		if (b->_b_waiters) __wake(&b->_b_lock, 1, 1);
 		a_store(&inst->last, 1);
 		if (inst->waiters)
-			__wake(&inst->last, -1, 0);
+			__wake(&inst->last, -1, 1);
 	} else {
 		a_store(&b->_b_lock, 0);
-		if (b->_b_waiters) __wake(&b->_b_lock, 1, 0);
-		__wait(&inst->last, &inst->waiters, 0, 0);
+		if (b->_b_waiters) __wake(&b->_b_lock, 1, 1);
+		__wait(&inst->last, &inst->waiters, 0, 1);
 	}
 
 	/* Last thread to exit the barrier wakes the instance owner */
 	if (a_fetch_add(&inst->count,-1)==1 && a_fetch_add(&inst->finished,1))
-		__wake(&inst->finished, 1, 0);
+		__wake(&inst->finished, 1, 1);
 
 	return 0;
 }
diff --git a/src/thread/pthread_barrierattr_setpshared.c b/src/thread/pthread_barrierattr_setpshared.c
index d6c7f9a3..1edc95dc 100644
--- a/src/thread/pthread_barrierattr_setpshared.c
+++ b/src/thread/pthread_barrierattr_setpshared.c
@@ -2,6 +2,6 @@
 
 int pthread_barrierattr_setpshared(pthread_barrierattr_t *a, int pshared)
 {
-	*a = !!pshared;
+	*a = pshared ? INT_MIN : 0;
 	return 0;
 }