rcu: Fix detection of abruptly-ending stall
Paul E. McKenney [Wed, 9 May 2012 15:45:12 +0000 (08:45 -0700)]
The code that attempts to identify stalls that end just as we detect
them is broken by both flavors of initialization failure.  This commit
therefore properly initializes and computes the count of the number
of reasons why the RCU grace period is stalled.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>

kernel/rcutree.c

index 4b97bba..dc8c528 100644 (file)
@@ -733,7 +733,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp)
        int cpu;
        long delta;
        unsigned long flags;
-       int ndetected;
+       int ndetected = 0;
        struct rcu_node *rnp = rcu_get_root(rsp);
 
        /* Only let one CPU complain about others per time interval. */
@@ -774,7 +774,7 @@ static void print_other_cpu_stall(struct rcu_state *rsp)
         */
        rnp = rcu_get_root(rsp);
        raw_spin_lock_irqsave(&rnp->lock, flags);
-       ndetected = rcu_print_task_stall(rnp);
+       ndetected += rcu_print_task_stall(rnp);
        raw_spin_unlock_irqrestore(&rnp->lock, flags);
 
        print_cpu_stall_info_end();