rcu: Optionally run grace-period kthreads at real-time priority

Recent testing has shown that under heavy load, running RCU's grace-period
kthreads at real-time priority can improve performance (according to 0day
test robot) and reduce the incidence of RCU CPU stall warnings.  However,
most systems do just fine with the default non-realtime priorities for
these kthreads, and it does not make sense to expose the entire user
base to any risk stemming from this change, given that this change is
of use only to a few users running extremely heavy workloads.

Therefore, this commit allows users to specify realtime priorities
for the grace-period kthreads, but leaves them running SCHED_OTHER
by default.  The realtime priority may be specified at build time
via the RCU_KTHREAD_PRIO Kconfig parameter, or at boot time via the
rcutree.kthread_prio parameter.  Either way, 0 says to continue the
default SCHED_OTHER behavior and values from 1-99 specify that priority
of SCHED_FIFO behavior.  Note that a value of 0 is not permitted when
the RCU_BOOST Kconfig parameter is specified.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

3 files changed, 27 insertions, 8 deletions

diff --git a/init/Kconfig b/init/Kconfig
index 9afb971..d3ee66a 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -668,9 +668,10 @@ config RCU_BOOST
 
 config RCU_KTHREAD_PRIO
 	int "Real-time priority to use for RCU worker threads"
-	range 1 99
-	depends on RCU_BOOST
-	default 1
+	range 1 99 if RCU_BOOST
+	range 0 99 if !RCU_BOOST
+	default 1 if RCU_BOOST
+	default 0 if !RCU_BOOST
 	help
 	  This option specifies the SCHED_FIFO priority value that will be
 	  assigned to the rcuc/n and rcub/n threads and is also the value
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 5987fdc..75ce123 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -156,6 +156,10 @@ static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu)
 static void invoke_rcu_core(void);
 static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp);
 
+/* rcuc/rcub kthread realtime priority */
+static int kthread_prio = CONFIG_RCU_KTHREAD_PRIO;
+module_param(kthread_prio, int, 0644);
+
 /*
  * Track the rcutorture test sequence number and the update version
  * number within a given test.  The rcutorture_testseq is incremented
@@ -3597,17 +3601,35 @@ static int rcu_pm_notify(struct notifier_block *self,
 static int __init rcu_spawn_gp_kthread(void)
 {
 	unsigned long flags;
+	int kthread_prio_in = kthread_prio;
 	struct rcu_node *rnp;
 	struct rcu_state *rsp;
+	struct sched_param sp;
 	struct task_struct *t;
 
+	/* Force priority into range. */
+	if (IS_ENABLED(CONFIG_RCU_BOOST) && kthread_prio < 1)
+		kthread_prio = 1;
+	else if (kthread_prio < 0)
+		kthread_prio = 0;
+	else if (kthread_prio > 99)
+		kthread_prio = 99;
+	if (kthread_prio != kthread_prio_in)
+		pr_alert("rcu_spawn_gp_kthread(): Limited prio to %d from %d\n",
+			 kthread_prio, kthread_prio_in);
+
 	rcu_scheduler_fully_active = 1;
 	for_each_rcu_flavor(rsp) {
-		t = kthread_run(rcu_gp_kthread, rsp, "%s", rsp->name);
+		t = kthread_create(rcu_gp_kthread, rsp, "%s", rsp->name);
 		BUG_ON(IS_ERR(t));
 		rnp = rcu_get_root(rsp);
 		raw_spin_lock_irqsave(&rnp->lock, flags);
 		rsp->gp_kthread = t;
+		if (kthread_prio) {
+			sp.sched_priority = kthread_prio;
+			sched_setscheduler_nocheck(t, SCHED_FIFO, &sp);
+		}
+		wake_up_process(t);
 		raw_spin_unlock_irqrestore(&rnp->lock, flags);
 	}
 	rcu_spawn_nocb_kthreads();
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
index 81ff8b9..dcb3263 100644
--- a/kernel/rcu/tree_plugin.h
+++ b/kernel/rcu/tree_plugin.h
@@ -34,10 +34,6 @@
 
 #include "../locking/rtmutex_common.h"
 
-/* rcuc/rcub kthread realtime priority */
-static int kthread_prio = CONFIG_RCU_KTHREAD_PRIO;
-module_param(kthread_prio, int, 0644);
-
 /*
  * Control variables for per-CPU and per-rcu_node kthreads.  These
  * handle all flavors of RCU.