Re: ANSI/IEEE POSIX 1003.1 and GCC (+ Other GNU)

[Richard Henderson]

On 01/07/2010 01:59 AM, Takis Psarogiannakopoulos wrote:

The right thing to do is use the depth (count above) of the mutex to
decide if the mutex is free. The owner shoud be valid only if the mutex is
locked.

I believe the following should fix this for you, at least for libgomp.
If you have a system for which the thread-id 0 is actually used, would
you mind testing it?


r~

diff --git a/libgomp/config/posix95/lock.c b/libgomp/config/posix95/lock.c
index 2242041..670302d 100644
--- a/libgomp/config/posix95/lock.c
+++ b/libgomp/config/posix95/lock.c
@@ -253,7 +253,7 @@ gomp_set_nest_lock_25 (omp_nest_lock_25_t *lock)
 {
   pthread_t me = pthread_self ();
 
-  if (lock->owner != me)
+  if (lock->count == 0 || lock->owner != me)
     {
       pthread_mutex_lock (&lock->lock);
       lock->owner = me;
@@ -279,7 +279,7 @@ gomp_test_nest_lock_25 (omp_nest_lock_25_t *lock)
 {
   pthread_t me = pthread_self ();
 
-  if (lock->owner != me)
+  if (lock->count == 0 || lock->owner != me)
     {
       if (pthread_mutex_trylock (&lock->lock) != 0)
        return 0;