Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <20221221170436.449941687@redhat.com>
User-Agent: quilt/0.66
Date:   Wed, 21 Dec 2022 13:58:07 -0300
From:   Marcelo Tosatti <mtosatti@redhat.com>
To:     atomlin@atomlin.com, frederic@kernel.org
Cc:     cl@linux.com, tglx@linutronix.de, mingo@kernel.org,
        peterz@infradead.org, pauld@redhat.com, neelx@redhat.com,
        oleksandr@natalenko.name, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org, Marcelo Tosatti <mtosatti@redhat.com>
Subject: [PATCH v11 6/6] mm/vmstat: avoid queueing work item if cpu stats are clean
References: <20221221165801.362118576@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Precedence: bulk

It is not necessary to queue work item to run refresh_vm_stats
on a remote CPU if that CPU has no dirty stats and no per-CPU
allocations for remote nodes.

This fixes sosreport hang (which uses vmstat_refresh) with
spinning SCHED_FIFO process.

Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>

Index: linux-2.6/mm/vmstat.c
===================================================================
--- linux-2.6.orig/mm/vmstat.c
+++ linux-2.6/mm/vmstat.c
@@ -1917,6 +1917,31 @@ static const struct seq_operations vmsta
 
 #ifdef CONFIG_SMP
 #ifdef CONFIG_PROC_FS
+static bool need_drain_remote_zones(int cpu)
+{
+#ifdef CONFIG_NUMA
+	struct zone *zone;
+
+	for_each_populated_zone(zone) {
+		struct per_cpu_pages *pcp;
+
+		pcp = per_cpu_ptr(zone->per_cpu_pageset, cpu);
+		if (!pcp->count)
+			continue;
+
+		if (!pcp->expire)
+			continue;
+
+		if (zone_to_nid(zone) == cpu_to_node(cpu))
+			continue;
+
+		return true;
+	}
+#endif
+
+	return false;
+}
+
 static void refresh_vm_stats(struct work_struct *work)
 {
 	refresh_cpu_vm_stats(true);
@@ -1926,8 +1951,12 @@ int vmstat_refresh(struct ctl_table *tab
 		   void *buffer, size_t *lenp, loff_t *ppos)
 {
 	long val;
-	int err;
-	int i;
+	int i, cpu;
+	struct work_struct __percpu *works;
+
+	works = alloc_percpu(struct work_struct);
+	if (!works)
+		return -ENOMEM;
 
 	/*
 	 * The regular update, every sysctl_stat_interval, may come later
@@ -1941,9 +1970,21 @@ int vmstat_refresh(struct ctl_table *tab
 	 * transiently negative values, report an error here if any of
 	 * the stats is negative, so we know to go looking for imbalance.
 	 */
-	err = schedule_on_each_cpu(refresh_vm_stats);
-	if (err)
-		return err;
+	cpus_read_lock();
+	for_each_online_cpu(cpu) {
+		struct work_struct *work = per_cpu_ptr(works, cpu);
+		struct vmstat_dirty *vms = per_cpu_ptr(&vmstat_dirty_pcpu, cpu);
+
+		INIT_WORK(work, refresh_vm_stats);
+
+		if (vms->dirty || need_drain_remote_zones(cpu))
+			schedule_work_on(cpu, work);
+	}
+	for_each_online_cpu(cpu)
+		flush_work(per_cpu_ptr(works, cpu));
+	cpus_read_unlock();
+	free_percpu(works);
+
 	for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++) {
 		/*
 		 * Skip checking stats known to go negative occasionally.