rcu: Prevent soft-lockup complaints about no-CBs CPUs
authorPaul Gortmaker <paul.gortmaker@windriver.com>
Thu, 20 Dec 2012 17:35:02 +0000 (09:35 -0800)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Tue, 8 Jan 2013 22:12:18 +0000 (14:12 -0800)
The wait_event() at the head of the rcu_nocb_kthread() can result in
soft-lockup complaints if the CPU in question does not register RCU
callbacks for an extended period.  This commit therefore changes
the wait_event() to a wait_event_interruptible().

Reported-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
kernel/rcutree_plugin.h

index f6e5ec2932b4aed0ff40016e8fdf77e5b9c1f366..43dba2d798ac7c3e867de88d85d08f6c86a45901 100644 (file)
@@ -2366,10 +2366,11 @@ static int rcu_nocb_kthread(void *arg)
        for (;;) {
                /* If not polling, wait for next batch of callbacks. */
                if (!rcu_nocb_poll)
-                       wait_event(rdp->nocb_wq, rdp->nocb_head);
+                       wait_event_interruptible(rdp->nocb_wq, rdp->nocb_head);
                list = ACCESS_ONCE(rdp->nocb_head);
                if (!list) {
                        schedule_timeout_interruptible(1);
+                       flush_signals(current);
                        continue;
                }