crypto: iaa - Optimize rebalance_wq_table()

JIRA: https://issues.redhat.com/browse/RHEL-95629
Upstream Status: merged into the linux.git

commit 714ca27
Author: Yury Norov <yury.norov@gmail.com>
Date:   Thu May 8 15:59:50 2025 -0400

    crypto: iaa - Optimize rebalance_wq_table()

    The function opencodes for_each_cpu() by using a plain for-loop. The
    loop calls cpumask_weight() inside the conditional section. Because
    cpumask_weight() is O(1), the overall complexity of the function is
    O(node * node_cpus^2). Also, cpumask_nth() internally calls hweight(),
    which, if not hardware accelerated, is slower than cpumask_next() in
    for_each_cpu().

    If switched to the dedicated for_each_cpu(), the rebalance_wq_table()
    can drop calling cpumask_weight(), together with some housekeeping code.
    This makes the overall complexity O(node * node_cpus), or simply speaking
    O(nr_cpu_ids).

    While there, fix opencoded for_each_possible_cpu() too.

    Signed-off-by: Yury Norov <yury.norov@gmail.com>
    Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>

Signed-off-by: Vladis Dronov <vdronov@redhat.com>

Author: nefigtut

drivers/crypto/intel/iaa/iaa_crypto_main.c

@@ -896,7 +896,7 @@ static int wq_table_add_wqs(int iaa, int cpu)
 static void rebalance_wq_table(void)
 {
 	const struct cpumask *node_cpus;
-	int node, cpu, iaa = -1;
+	int node_cpu, node, cpu, iaa = 0;
 
 	if (nr_iaa == 0)
 		return;
@@ -907,36 +907,29 @@ static void rebalance_wq_table(void)
 	clear_wq_table();
 
 	if (nr_iaa == 1) {
-		for (cpu = 0; cpu < nr_cpus; cpu++) {
-			if (WARN_ON(wq_table_add_wqs(0, cpu))) {
-				pr_debug("could not add any wqs for iaa 0 to cpu %d!\n", cpu);
-				return;
-			}
+		for_each_possible_cpu(cpu) {
+			if (WARN_ON(wq_table_add_wqs(0, cpu)))
+				goto err;
 		}
 
 		return;
 	}
 
 	for_each_node_with_cpus(node) {
+		cpu = 0;
 		node_cpus = cpumask_of_node(node);
 
-		for (cpu = 0; cpu <  cpumask_weight(node_cpus); cpu++) {
-			int node_cpu = cpumask_nth(cpu, node_cpus);
-
-			if (WARN_ON(node_cpu >= nr_cpu_ids)) {
-				pr_debug("node_cpu %d doesn't exist!\n", node_cpu);
-				return;
-			}
-
-			if ((cpu % cpus_per_iaa) == 0)
-				iaa++;
-
-			if (WARN_ON(wq_table_add_wqs(iaa, node_cpu))) {
-				pr_debug("could not add any wqs for iaa %d to cpu %d!\n", iaa, cpu);
-				return;
-			}
+		for_each_cpu(node_cpu, node_cpus) {
+			iaa = cpu / cpus_per_iaa;
+			if (WARN_ON(wq_table_add_wqs(iaa, node_cpu)))
+				goto err;
+			cpu++;
 		}
 	}
+
+	return;
+err:
+	pr_debug("could not add any wqs for iaa %d to cpu %d!\n", iaa, cpu);
 }
 
 static inline int check_completion(struct device *dev,