refscale: Allow summarization of verbose output

The refscale test prints enough per-kthread console output to provoke RCU
CPU stall warnings on large systems.  This commit therefore allows this
output to be summarized.  For example, the refscale.verbose_batched=32
boot parameter would causes only every 32nd line of output to be logged.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>

Documentation/admin-guide/kernel-parameters.txt

@@ -4557,6 +4557,12 @@
 	refscale.verbose= [KNL]
 			Enable additional printk() statements.
 
+	refscale.verbose_batched= [KNL]
+			Batch the additional printk() statements.  If zero
+			(the default) or negative, print everything.  Otherwise,
+			print every Nth verbose statement, where N is the value
+			specified.
+
 	relax_domain_level=
 			[KNL, SMP] Set scheduler's default relax_domain_level.
 			See Documentation/admin-guide/cgroup-v1/cpusets.rst.

kernel/rcu/refscale.c

@@ -46,6 +46,16 @@
 #define VERBOSE_SCALEOUT(s, x...) \
 	do { if (verbose) pr_alert("%s" SCALE_FLAG s, scale_type, ## x); } while (0)
 
+static atomic_t verbose_batch_ctr;
+
+#define VERBOSE_SCALEOUT_BATCH(s, x...)							\
+do {											\
+	if (verbose &&									\
+	    (verbose_batched <= 0 ||							\
+	     !(atomic_inc_return(&verbose_batch_ctr) % verbose_batched)))		\
+		pr_alert("%s" SCALE_FLAG s, scale_type, ## x);				\
+} while (0)
+
 #define VERBOSE_SCALEOUT_ERRSTRING(s, x...) \
 	do { if (verbose) pr_alert("%s" SCALE_FLAG "!!! " s, scale_type, ## x); } while (0)
 
@@ -57,6 +67,7 @@ module_param(scale_type, charp, 0444);
 MODULE_PARM_DESC(scale_type, "Type of test (rcu, srcu, refcnt, rwsem, rwlock.");
 
 torture_param(int, verbose, 0, "Enable verbose debugging printk()s");
+torture_param(int, verbose_batched, 0, "Batch verbose debugging printk()s");
 
 // Wait until there are multiple CPUs before starting test.
 torture_param(int, holdoff, IS_BUILTIN(CONFIG_RCU_REF_SCALE_TEST) ? 10 : 0,
@@ -368,14 +379,14 @@ ref_scale_reader(void *arg)
 	u64 start;
 	s64 duration;
 
-	VERBOSE_SCALEOUT("ref_scale_reader %ld: task started", me);
+	VERBOSE_SCALEOUT_BATCH("ref_scale_reader %ld: task started", me);
 	set_cpus_allowed_ptr(current, cpumask_of(me % nr_cpu_ids));
 	set_user_nice(current, MAX_NICE);
 	atomic_inc(&n_init);
 	if (holdoff)
 		schedule_timeout_interruptible(holdoff * HZ);
 repeat:
-	VERBOSE_SCALEOUT("ref_scale_reader %ld: waiting to start next experiment on cpu %d", me, smp_processor_id());
+	VERBOSE_SCALEOUT_BATCH("ref_scale_reader %ld: waiting to start next experiment on cpu %d", me, smp_processor_id());
 
 	// Wait for signal that this reader can start.
 	wait_event(rt->wq, (atomic_read(&nreaders_exp) && smp_load_acquire(&rt->start_reader)) ||
@@ -392,7 +403,7 @@ ref_scale_reader(void *arg)
 		while (atomic_read_acquire(&n_started))
 			cpu_relax();
 
-	VERBOSE_SCALEOUT("ref_scale_reader %ld: experiment %d started", me, exp_idx);
+	VERBOSE_SCALEOUT_BATCH("ref_scale_reader %ld: experiment %d started", me, exp_idx);
 
 
 	// To reduce noise, do an initial cache-warming invocation, check
@@ -421,8 +432,8 @@ ref_scale_reader(void *arg)
 	if (atomic_dec_and_test(&nreaders_exp))
 		wake_up(&main_wq);
 
-	VERBOSE_SCALEOUT("ref_scale_reader %ld: experiment %d ended, (readers remaining=%d)",
-			me, exp_idx, atomic_read(&nreaders_exp));
+	VERBOSE_SCALEOUT_BATCH("ref_scale_reader %ld: experiment %d ended, (readers remaining=%d)",
+				me, exp_idx, atomic_read(&nreaders_exp));
 
 	if (!torture_must_stop())
 		goto repeat;