rcu: avoid hammering sched with yet another bound RT kthread
authorPaul E. McKenney <paul.mckenney@linaro.org>
Mon, 18 Apr 2011 06:45:23 +0000 (23:45 -0700)
committerPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Fri, 6 May 2011 06:16:55 +0000 (23:16 -0700)
The scheduler does not appear to take kindly to having multiple
real-time threads bound to a CPU that is going offline.  So this
commit is a temporary hack-around to avoid that happening.

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

index c224da41890cd3382b060ffb1567d3827f71e21c..4d256dba7d9b0d62d45f7c81b74eb2cb755be6e2 100644 (file)
@@ -163,11 +163,11 @@ static int stutter_pause_test;
 #endif
 int rcutorture_runnable = RCUTORTURE_RUNNABLE_INIT;
 
-#ifdef CONFIG_RCU_BOOST
+#if defined(CONFIG_RCU_BOOST) && !defined(CONFIG_HOTPLUG_CPU)
 #define rcu_can_boost() 1
-#else /* #ifdef CONFIG_RCU_BOOST */
+#else /* #if defined(CONFIG_RCU_BOOST) && !defined(CONFIG_HOTPLUG_CPU) */
 #define rcu_can_boost() 0
-#endif /* #else #ifdef CONFIG_RCU_BOOST */
+#endif /* #else #if defined(CONFIG_RCU_BOOST) && !defined(CONFIG_HOTPLUG_CPU) */
 
 static unsigned long boost_starttime;  /* jiffies of next boost test start. */
 DEFINE_MUTEX(boost_mutex);             /* protect setting boost_starttime */