mutexes: Give more informative mutex warning in the !lock->owner case
authorChuansheng Liu <chuansheng.liu@intel.com>
Wed, 4 Dec 2013 05:58:13 +0000 (13:58 +0800)
committerIngo Molnar <mingo@kernel.org>
Tue, 17 Dec 2013 14:35:10 +0000 (15:35 +0100)
When mutex debugging is enabled and an imbalanced mutex_unlock()
is called, we get the following, slightly confusing warning:

  [  364.208284] DEBUG_LOCKS_WARN_ON(lock->owner != current)

But in that case the warning is due to an imbalanced mutex_unlock() call,
and the lock->owner is NULL - so the message is misleading.

So improve the message by testing for this case specifically:

   DEBUG_LOCKS_WARN_ON(!lock->owner)

Signed-off-by: Liu, Chuansheng <chuansheng.liu@intel.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Link: http://lkml.kernel.org/r/1386136693.3650.48.camel@cliu38-desktop-build
[ Improved the changelog, changed the patch to use !lock->owner consistently. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
kernel/locking/mutex-debug.c

index 7e3443fe1f48a1dbea11f7287fd3a362f1adb2a7..faf6f5b53e775eefbc7ee4ea734ff0998be68d01 100644 (file)
@@ -75,7 +75,12 @@ void debug_mutex_unlock(struct mutex *lock)
                return;
 
        DEBUG_LOCKS_WARN_ON(lock->magic != lock);
-       DEBUG_LOCKS_WARN_ON(lock->owner != current);
+
+       if (!lock->owner)
+               DEBUG_LOCKS_WARN_ON(!lock->owner);
+       else
+               DEBUG_LOCKS_WARN_ON(lock->owner != current);
+
        DEBUG_LOCKS_WARN_ON(!lock->wait_list.prev && !lock->wait_list.next);
        mutex_clear_owner(lock);
 }