rcu: Pull rcu_qs_ctr into rcu_dynticks structure

The rcu_qs_ctr variable is yet another isolated per-CPU variable,
so this commit pulls it into the pre-existing rcu_dynticks per-CPU
structure.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

Documentation/RCU/Design/Data-Structures/Data-Structures.html

@@ -1105,6 +1105,7 @@ Its fields are as follows:
   2   int dynticks_nmi_nesting;
   3   atomic_t dynticks;
   4   int rcu_sched_qs_mask;
+  5   unsigned long rcu_qs_ctr;
 </pre>
 
 <p>The <tt>-&gt;dynticks_nesting</tt> field counts the
@@ -1123,12 +1124,19 @@ CPU's transitions to and from dyntick-idle mode, so that this counter
 has an even value when the CPU is in dyntick-idle mode and an odd
 value otherwise.
 
-</p><p>Finally, the  <tt>-&gt;rcu_sched_qs_mask</tt> field is used
+</p><p>The <tt>-&gt;rcu_sched_qs_mask</tt> field is used
 to record the fact that the RCU core code would really like to
-see a quiescent state from the corresponding CPU.
+see a quiescent state from the corresponding CPU, so much so that
+it is willing to call for heavy-weight dyntick-counter operations.
 This flag is checked by RCU's context-switch and <tt>cond_resched()</tt>
 code, which provide a momentary idle sojourn in response.
 
+</p><p>Finally the <tt>-&gt;rcu_qs_ctr</tt> field is used to record
+quiescent states from <tt>cond_resched()</tt>.
+Because <tt>cond_resched()</tt> can execute quite frequently, this
+must be quite lightweight, as in a non-atomic increment of this
+per-CPU field.
+
 <table>
 <tr><th>&nbsp;</th></tr>
 <tr><th align="left">Quick Quiz:</th></tr>

kernel/rcu/tree.c

@@ -436,9 +436,6 @@ bool rcu_eqs_special_set(int cpu)
 	return true;
 }
 
-DEFINE_PER_CPU_SHARED_ALIGNED(unsigned long, rcu_qs_ctr);
-EXPORT_PER_CPU_SYMBOL_GPL(rcu_qs_ctr);
-
 /*
  * Let the RCU core know that this CPU has gone through the scheduler,
  * which is a quiescent state.  This is called when the need for a
@@ -542,7 +539,7 @@ void rcu_all_qs(void)
 		rcu_sched_qs();
 		preempt_enable();
 	}
-	this_cpu_inc(rcu_qs_ctr);
+	this_cpu_inc(rcu_dynticks.rcu_qs_ctr);
 	barrier(); /* Avoid RCU read-side critical sections leaking up. */
 }
 EXPORT_SYMBOL_GPL(rcu_all_qs);
@@ -1315,7 +1312,7 @@ static int rcu_implicit_dynticks_qs(struct rcu_data *rdp,
 	 */
 	rnp = rdp->mynode;
 	if (time_after(jiffies, rdp->rsp->gp_start + jtsq) &&
-	    READ_ONCE(rdp->rcu_qs_ctr_snap) != per_cpu(rcu_qs_ctr, rdp->cpu) &&
+	    READ_ONCE(rdp->rcu_qs_ctr_snap) != per_cpu(rcu_dynticks.rcu_qs_ctr, rdp->cpu) &&
 	    READ_ONCE(rdp->gpnum) == rnp->gpnum && !rdp->gpwrap) {
 		trace_rcu_fqs(rdp->rsp->name, rdp->gpnum, rdp->cpu, TPS("rqc"));
 		return 1;
@@ -2024,7 +2021,7 @@ static bool __note_gp_changes(struct rcu_state *rsp, struct rcu_node *rnp,
 		trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpustart"));
 		need_gp = !!(rnp->qsmask & rdp->grpmask);
 		rdp->cpu_no_qs.b.norm = need_gp;
-		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr);
+		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_dynticks.rcu_qs_ctr);
 		rdp->core_needs_qs = need_gp;
 		zero_cpu_stall_ticks(rdp);
 		WRITE_ONCE(rdp->gpwrap, false);
@@ -2622,7 +2619,7 @@ rcu_report_qs_rdp(int cpu, struct rcu_state *rsp, struct rcu_data *rdp)
 		 * within the current grace period.
 		 */
 		rdp->cpu_no_qs.b.norm = true;	/* need qs for new gp. */
-		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_qs_ctr);
+		rdp->rcu_qs_ctr_snap = __this_cpu_read(rcu_dynticks.rcu_qs_ctr);
 		raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
 		return;
 	}
@@ -3620,7 +3617,7 @@ static int __rcu_pending(struct rcu_state *rsp, struct rcu_data *rdp)
 	/* Is the RCU core waiting for a quiescent state from this CPU? */
 	if (rcu_scheduler_fully_active &&
 	    rdp->core_needs_qs && rdp->cpu_no_qs.b.norm &&
-	    rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_qs_ctr)) {
+	    rdp->rcu_qs_ctr_snap == __this_cpu_read(rcu_dynticks.rcu_qs_ctr)) {
 		rdp->n_rp_core_needs_qs++;
 	} else if (rdp->core_needs_qs && !rdp->cpu_no_qs.b.norm) {
 		rdp->n_rp_report_qs++;
@@ -3933,7 +3930,7 @@ rcu_init_percpu_data(int cpu, struct rcu_state *rsp)
 	rdp->gpnum = rnp->completed; /* Make CPU later note any new GP. */
 	rdp->completed = rnp->completed;
 	rdp->cpu_no_qs.b.norm = true;
-	rdp->rcu_qs_ctr_snap = per_cpu(rcu_qs_ctr, cpu);
+	rdp->rcu_qs_ctr_snap = per_cpu(rcu_dynticks.rcu_qs_ctr, cpu);
 	rdp->core_needs_qs = false;
 	trace_rcu_grace_period(rsp->name, rdp->gpnum, TPS("cpuonl"));
 	raw_spin_unlock_irqrestore_rcu_node(rnp, flags);

kernel/rcu/tree.h

@@ -113,7 +113,8 @@ struct rcu_dynticks {
 				    /* Process level is worth LLONG_MAX/2. */
 	int dynticks_nmi_nesting;   /* Track NMI nesting level. */
 	atomic_t dynticks;	    /* Even value for idle, else odd. */
-	int rcu_sched_qs_mask;      /* GP old, need quiescent state. */
+	int rcu_sched_qs_mask;      /* GP old, need heavy quiescent state. */
+	unsigned long rcu_qs_ctr;   /* Light universal quiescent state ctr. */
 #ifdef CONFIG_NO_HZ_FULL_SYSIDLE
 	long long dynticks_idle_nesting;
 				    /* irq/process nesting level from idle. */

kernel/rcu/tree_trace.c

@@ -45,8 +45,6 @@
 #define RCU_TREE_NONCORE
 #include "tree.h"
 
-DECLARE_PER_CPU_SHARED_ALIGNED(unsigned long, rcu_qs_ctr);
-
 static int r_open(struct inode *inode, struct file *file,
 					const struct seq_operations *op)
 {
@@ -121,7 +119,7 @@ static void print_one_rcu_data(struct seq_file *m, struct rcu_data *rdp)
 		   cpu_is_offline(rdp->cpu) ? '!' : ' ',
 		   ulong2long(rdp->completed), ulong2long(rdp->gpnum),
 		   rdp->cpu_no_qs.b.norm,
-		   rdp->rcu_qs_ctr_snap == per_cpu(rcu_qs_ctr, rdp->cpu),
+		   rdp->rcu_qs_ctr_snap == per_cpu(rdp->dynticks->rcu_qs_ctr, rdp->cpu),
 		   rdp->core_needs_qs);
 	seq_printf(m, " dt=%d/%llx/%d df=%lu",
 		   rcu_dynticks_snap(rdp->dynticks),