It seems to me that the schedule() issued  after the wakeup in locks.c may
be a nono. To give it more sense the task should go in the YIELD
scheduler policy so lock/unlock between task will scale better.

Index: fs//locks.c
===================================================================
RCS file: /var/cvs/linux/fs/locks.c,v
retrieving revision 1.1.2.3
diff -u -r1.1.2.3 locks.c
--- locks.c	1999/04/09 16:53:45	1.1.2.3
+++ linux/fs/locks.c	1999/05/11 00:39:20
@@ -267,6 +267,7 @@
 			/* Let the blocked process remove waiter from the
 			 * block list when it gets scheduled.
 			 */
+			current->policy |= SCHED_YIELD;
 			schedule();
 		else
 			/* Remove waiter from the block list, because by the

Andrea Arcangeli


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/