rcu: further lower priority in rcu_yield()
authorPaul E. McKenney <paul.mckenney@linaro.org>
Sun, 27 Mar 2011 05:01:35 +0000 (22:01 -0700)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Fri, 6 May 2011 06:16:59 +0000 (23:16 -0700)
Although rcu_yield() dropped from real-time to normal priority, there
is always the possibility that the competing tasks have been niced.
So nice to 19 in rcu_yield() to help ensure that other tasks have a
better chance of running.

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 2c07adb..fe85600 100644 (file)
@@ -1561,6 +1561,7 @@ static void rcu_yield(void (*f)(unsigned long), unsigned long arg)
        mod_timer(&yield_timer, jiffies + 2);
        sp.sched_priority = 0;
        sched_setscheduler_nocheck(current, SCHED_NORMAL, &sp);
+       set_user_nice(current, 19);
        schedule();
        sp.sched_priority = RCU_KTHREAD_PRIO;
        sched_setscheduler_nocheck(current, SCHED_FIFO, &sp);