rcu: Latch normal synchronize_rcu() path on flood

Currently, rcu_normal_wake_from_gp is only enabled by default
on small systems(<= 16 CPUs) or when a user explicitly set it
enabled.

Introduce an adaptive latching mechanism:
 * Track the number of in-flight synchronize_rcu() requests
   using a new rcu_sr_normal_count counter;

 * If the count reaches/exceeds RCU_SR_NORMAL_LATCH_THR(64),
   it sets the rcu_sr_normal_latched, reverting new requests
   onto the scaled wait_rcu_gp() path;

 * The latch is cleared only when the pending requests are fully
   drained(nr == 0);

 * Enables rcu_normal_wake_from_gp by default for all systems,
   relying on this dynamic throttling instead of static CPU
   limits.

Testing(synthetic flood workload):
  * Kernel version: 6.19.0-rc6
  * Number of CPUs: 1536
  * 60K concurrent synchronize_rcu() calls

Perf(cycles, system-wide):
  total cycles: 932020263832
  rcu_sr_normal_add_req(): 2650282811 cycles(~0.28%)

Perf report excerpt:
  0.01%  0.01%  sync_test/...  [k] rcu_sr_normal_add_req

Measured overhead of rcu_sr_normal_add_req() remained ~0.28%
of total CPU cycles in this synthetic stress test.

Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Tested-by: Samir M <samir@linux.ibm.com>
Suggested-by: Joel Fernandes <joelagnelf@nvidia.com>
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>

2 files changed, 44 insertions, 18 deletions

diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 4d0f545fb3ec..d5db2e85d551 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5862,13 +5862,13 @@ Kernel parameters
 			use a call_rcu[_hurry]() path. Please note, this is for a
 			normal grace period.
 
-			How to enable it:
+			How to disable it:
 
-			echo 1 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp
-			or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=1"
+			echo 0 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp
+			or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=0"
 
-			Default is 1 if num_possible_cpus() <= 16 and it is not explicitly
-			disabled by the boot parameter passing 0.
+			Default is 1 if it is not explicitly disabled by the boot parameter
+			passing 0.
 
 	rcuscale.gp_async= [KNL]
 			Measure performance of asynchronous
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 09f0cef5014c..afb9e7db8f78 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -1632,17 +1632,21 @@ static void rcu_sr_put_wait_head(struct llist_node *node)
 	atomic_set_release(&sr_wn->inuse, 0);
 }
 
-/* Enable rcu_normal_wake_from_gp automatically on small systems. */
-#define WAKE_FROM_GP_CPU_THRESHOLD 16
-
-static int rcu_normal_wake_from_gp = -1;
+static int rcu_normal_wake_from_gp = 1;
 module_param(rcu_normal_wake_from_gp, int, 0644);
 static struct workqueue_struct *sync_wq;
 
+#define RCU_SR_NORMAL_LATCH_THR 64
+
+/* Number of in-flight synchronize_rcu() calls queued on srs_next. */
+static atomic_long_t rcu_sr_normal_count;
+static int rcu_sr_normal_latched; /* 0/1 */
+
 static void rcu_sr_normal_complete(struct llist_node *node)
 {
 	struct rcu_synchronize *rs = container_of(
 		(struct rcu_head *) node, struct rcu_synchronize, head);
+	long nr;
 
 	WARN_ONCE(IS_ENABLED(CONFIG_PROVE_RCU) &&
 		!poll_state_synchronize_rcu_full(&rs->oldstate),
@@ -1650,6 +1654,15 @@ static void rcu_sr_normal_complete(struct llist_node *node)
 
 	/* Finally. */
 	complete(&rs->completion);
+	nr = atomic_long_dec_return(&rcu_sr_normal_count);
+	WARN_ON_ONCE(nr < 0);
+
+	/*
+	 * Unlatch: switch back to normal path when fully
+	 * drained and if it has been latched.
+	 */
+	if (nr == 0)
+		(void)cmpxchg_relaxed(&rcu_sr_normal_latched, 1, 0);
 }
 
 static void rcu_sr_normal_gp_cleanup_work(struct work_struct *work)
@@ -1795,6 +1808,24 @@ static bool rcu_sr_normal_gp_init(void)
 
 static void rcu_sr_normal_add_req(struct rcu_synchronize *rs)
 {
+	/*
+	 * Increment before publish to avoid a complete
+	 * vs enqueue race on latch.
+	 */
+	long nr = atomic_long_inc_return(&rcu_sr_normal_count);
+
+	/*
+	 * Latch when threshold is reached. Checking for an exact match
+	 * restricts cmpxchg() to a single context.
+	 *
+	 * This latch is intentionally relaxed and best-effort. Concurrent
+	 * set/clear can race and temporarily lose the latch, which is OK
+	 * because it only selects between the fast and fallback paths.
+	 */
+	if (nr == RCU_SR_NORMAL_LATCH_THR)
+		(void)cmpxchg_relaxed(&rcu_sr_normal_latched, 0, 1);
+
+	/* Publish for the GP kthread/worker. */
 	llist_add((struct llist_node *) &rs->head, &rcu_state.srs_next);
 }
 
@@ -3278,14 +3309,15 @@ static void synchronize_rcu_normal(void)
 {
 	struct rcu_synchronize rs;
 
+	init_rcu_head_on_stack(&rs.head);
 	trace_rcu_sr_normal(rcu_state.name, &rs.head, TPS("request"));
 
-	if (READ_ONCE(rcu_normal_wake_from_gp) < 1) {
+	if (READ_ONCE(rcu_normal_wake_from_gp) < 1 ||
+			READ_ONCE(rcu_sr_normal_latched)) {
 		wait_rcu_gp(call_rcu_hurry);
 		goto trace_complete_out;
 	}
 
-	init_rcu_head_on_stack(&rs.head);
 	init_completion(&rs.completion);
 
 	/*
@@ -3302,10 +3334,10 @@ static void synchronize_rcu_normal(void)
 
 	/* Now we can wait. */
 	wait_for_completion(&rs.completion);
-	destroy_rcu_head_on_stack(&rs.head);
 
 trace_complete_out:
 	trace_rcu_sr_normal(rcu_state.name, &rs.head, TPS("complete"));
+	destroy_rcu_head_on_stack(&rs.head);
 }
 
 /**
@@ -4904,12 +4936,6 @@ void __init rcu_init(void)
 	sync_wq = alloc_workqueue("sync_wq", WQ_MEM_RECLAIM | WQ_UNBOUND, 0);
 	WARN_ON(!sync_wq);
 
-	/* Respect if explicitly disabled via a boot parameter. */
-	if (rcu_normal_wake_from_gp < 0) {
-		if (num_possible_cpus() <= WAKE_FROM_GP_CPU_THRESHOLD)
-			rcu_normal_wake_from_gp = 1;
-	}
-
 	/* Fill in default value for rcutree.qovld boot parameter. */
 	/* -After- the rcu_node ->lock fields are initialized! */
 	if (qovld < 0)