rcu: Add check for CPUs going offline with callbacks queued
authorPaul E. McKenney <paul.mckenney@linaro.org>
Thu, 21 Jun 2012 18:26:42 +0000 (11:26 -0700)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Mon, 2 Jul 2012 19:34:25 +0000 (12:34 -0700)
If a CPU goes offline with callbacks queued, those callbacks might be
indefinitely postponed, which can result in a system hang.  This commit
therefore inserts warnings for this condition.

Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
kernel/rcutree.c

index e000a623e6358a93801c9e1bf0cb8d63b796d1b2..95c7b61e77e4c9d114b2725fe0faf7a3f2cc5fd0 100644 (file)
@@ -1509,6 +1509,9 @@ static void rcu_cleanup_dead_cpu(int cpu, struct rcu_state *rsp)
                raw_spin_unlock_irqrestore(&rnp->lock, flags);
        if (need_report & RCU_OFL_TASKS_EXP_GP)
                rcu_report_exp_rnp(rsp, rnp, true);
+       WARN_ONCE(rdp->qlen != 0 || rdp->nxtlist != NULL,
+                 "rcu_cleanup_dead_cpu: Callbacks on offline CPU %d: qlen=%lu, nxtlist=%p\n",
+                 cpu, rdp->qlen, rdp->nxtlist);
 }
 
 #else /* #ifdef CONFIG_HOTPLUG_CPU */