mm/demotion: add pg_data_t member to track node memory tier details

Also update different helpes to use NODE_DATA()->memtier.  Since node
specific memtier can change based on the reassignment of NUMA node to a
different memory tiers, accessing NODE_DATA()->memtier needs to happen
under an rcu read lock or memory_tier_lock.

Link: https://lkml.kernel.org/r/20220818131042.113280-7-aneesh.kumar@linux.ibm.com
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Acked-by: Wei Xu <weixugc@google.com>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Bharata B Rao <bharata@amd.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Hesham Almatary <hesham.almatary@huawei.com>
Cc: Jagdish Gediya <jvgediya.oss@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tim Chen <tim.c.chen@intel.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: SeongJae Park <sj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

include/linux/mmzone.h

@@ -1246,6 +1246,9 @@ typedef struct pglist_data {
 	/* Per-node vmstats */
 	struct per_cpu_nodestat __percpu *per_cpu_nodestats;
 	atomic_long_t		vm_stat[NR_VM_NODE_STAT_ITEMS];
+#ifdef CONFIG_NUMA
+	struct memory_tier __rcu *memtier;
+#endif
 } pg_data_t;
 
 #define node_present_pages(nid)	(NODE_DATA(nid)->node_present_pages)

mm/memory-tiers.c

@@ -4,6 +4,7 @@
 #include <linux/sysfs.h>
 #include <linux/kobject.h>
 #include <linux/memory.h>
+#include <linux/mmzone.h>
 #include <linux/memory-tiers.h>
 
 #include "internal.h"
@@ -141,12 +142,18 @@ static struct memory_tier *find_create_memory_tier(struct memory_dev_type *memty
 
 static struct memory_tier *__node_get_memory_tier(int node)
 {
-	struct memory_dev_type *memtype;
+	pg_data_t *pgdat;
 
-	memtype = node_memory_types[node];
+	pgdat = NODE_DATA(node);
-	if (memtype && node_isset(node, memtype->nodes))
+	if (!pgdat)
-		return memtype->memtier;
+		return NULL;
-	return NULL;
+	/*
+	 * Since we hold memory_tier_lock, we can avoid
+	 * RCU read locks when accessing the details. No
+	 * parallel updates are possible here.
+	 */
+	return rcu_dereference_check(pgdat->memtier,
+				     lockdep_is_held(&memory_tier_lock));
 }
 
 #ifdef CONFIG_MIGRATION
@@ -309,6 +316,8 @@ static struct memory_tier *set_node_memory_tier(int node)
 {
 	struct memory_tier *memtier;
 	struct memory_dev_type *memtype;
+	pg_data_t *pgdat = NODE_DATA(node);
+
 
 	lockdep_assert_held_once(&memory_tier_lock);
 
@@ -320,24 +329,45 @@ static struct memory_tier *set_node_memory_tier(int node)
 	memtype = node_memory_types[node].memtype;
 	node_set(node, memtype->nodes);
 	memtier = find_create_memory_tier(memtype);
+	if (!IS_ERR(memtier))
+		rcu_assign_pointer(pgdat->memtier, memtier);
 	return memtier;
 }
 
 static void destroy_memory_tier(struct memory_tier *memtier)
 {
 	list_del(&memtier->list);
+	/*
+	 * synchronize_rcu in clear_node_memory_tier makes sure
+	 * we don't have rcu access to this memory tier.
+	 */
 	kfree(memtier);
 }
 
 static bool clear_node_memory_tier(int node)
 {
 	bool cleared = false;
+	pg_data_t *pgdat;
 	struct memory_tier *memtier;
 
+	pgdat = NODE_DATA(node);
+	if (!pgdat)
+		return false;
+
+	/*
+	 * Make sure that anybody looking at NODE_DATA who finds
+	 * a valid memtier finds memory_dev_types with nodes still
+	 * linked to the memtier. We achieve this by waiting for
+	 * rcu read section to finish using synchronize_rcu.
+	 * This also enables us to free the destroyed memory tier
+	 * with kfree instead of kfree_rcu
+	 */
 	memtier = __node_get_memory_tier(node);
 	if (memtier) {
 		struct memory_dev_type *memtype;
 
+		rcu_assign_pointer(pgdat->memtier, NULL);
+		synchronize_rcu();
 		memtype = node_memory_types[node].memtype;
 		node_clear(node, memtype->nodes);
 		if (nodes_empty(memtype->nodes)) {