mm: vmalloc: improve description of vmap node layer

This patch adds extra explanation of recently added vmap node layer based
on community feedback. No functional change.

Link: https://lkml.kernel.org/r/[e-mail]
Signed-off-by: Uladzislau Rezki (Sony) <[e-mail]>
Reviewed-by: Lorenzo Stoakes <[e-mail]>
Cc: Baoquan He <[e-mail]>
Cc: Christoph Hellwig <[e-mail]>
Cc: Dave Chinner <[e-mail]>
Cc: Matthew Wilcox (Oracle) <[e-mail]>
Cc: Oleksiy Avramchenko <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>

mm/vmalloc.c

···
 static DEFINE_PER_CPU(struct vmap_area *, ne_fit_preload_node);
 
 /*
- * An effective vmap-node logic. Users make use of nodes instead
- * of a global heap. It allows to balance an access and mitigate
- * contention.
+ * This structure defines a single, solid model where a list and
+ * rb-tree are part of one entity protected by the lock. Nodes are
+ * sorted in ascending order, thus for O(1) access to left/right
+ * neighbors a list is used as well as for sequential traversal.
  */
 struct rb_list {
 	struct rb_root root;
···
 	spinlock_t lock;
 };
 
+/*
+ * A fast size storage contains VAs up to 1M size. A pool consists
+ * of linked between each other ready to go VAs of certain sizes.
+ * An index in the pool-array corresponds to number of pages + 1.
+ */
+#define MAX_VA_SIZE_PAGES 256
+
 struct vmap_pool {
 	struct list_head head;
 	unsigned long len;
 };
 
 /*
- * A fast size storage contains VAs up to 1M size.
+ * An effective vmap-node logic. Users make use of nodes instead
+ * of a global heap. It allows to balance an access and mitigate
+ * contention.
  */
-#define MAX_VA_SIZE_PAGES 256
-
 static struct vmap_node {
 	/* Simple size segregated storage. */
 	struct vmap_pool pool[MAX_VA_SIZE_PAGES];
···
 	unsigned long nr_purged;
 } single;
 
+/*
+ * Initial setup consists of one single node, i.e. a balancing
+ * is fully disabled. Later on, after vmap is initialized these
+ * parameters are updated based on a system capacity.
+ */
 static struct vmap_node *vmap_nodes = &single;
 static __read_mostly unsigned int nr_vmap_nodes = 1;
 static __read_mostly unsigned int vmap_zone_size = 1;
···
 			}
 		}
 
-		/* Attach the pool back if it has been partly decayed. */
+		/*
+		 * Attach the pool back if it has been partly decayed.
+		 * Please note, it is supposed that nobody(other contexts)
+		 * can populate the pool therefore a simple list replace
+		 * operation takes place here.
+		 */
 		if (!full_decay && !list_empty(&tmp_list)) {
 			spin_lock(&vn->pool_lock);
 			list_replace_init(&tmp_list, &vn->pool[i].head);
···
 	 * An addr_to_node_id(addr) converts an address to a node index
 	 * where a VA is located. If VA spans several zones and passed
 	 * addr is not the same as va->va_start, what is not common, we
-	 * may need to scan an extra nodes. See an example:
+	 * may need to scan extra nodes. See an example:
 	 *
-	 *      <--va-->
+	 *      <----va---->
 	 * -|-----|-----|-----|-----|-
 	 *     1     2     0     1
 	 *
-	 * VA resides in node 1 whereas it spans 1 and 2. If passed
-	 * addr is within a second node we should do extra work. We
-	 * should mention that it is rare and is a corner case from
-	 * the other hand it has to be covered.
+	 * VA resides in node 1 whereas it spans 1, 2 an 0. If passed
+	 * addr is within 2 or 0 nodes we should do extra work.
 	 */
 	i = j = addr_to_node_id(addr);
 	do {
···
 	struct vmap_area *va;
 	int i, j;
 
+	/*
+	 * Check the comment in the find_vmap_area() about the loop.
+	 */
 	i = j = addr_to_node_id(addr);
 	do {
 		vn = &vmap_nodes[i];
···
 	int i, n;
 
 #if BITS_PER_LONG == 64
-	/* A high threshold of max nodes is fixed and bound to 128. */
+	/*
+	 * A high threshold of max nodes is fixed and bound to 128,
+	 * thus a scale factor is 1 for systems where number of cores
+	 * are less or equal to specified threshold.
+	 *
+	 * As for NUMA-aware notes. For bigger systems, for example
+	 * NUMA with multi-sockets, where we can end-up with thousands
+	 * of cores in total, a "sub-numa-clustering" should be added.
+	 *
+	 * In this case a NUMA domain is considered as a single entity
+	 * with dedicated sub-nodes in it which describe one group or
+	 * set of cores. Therefore a per-domain purging is supposed to
+	 * be added as well as a per-domain balancing.
+	 */
 	n = clamp_t(unsigned int, num_possible_cpus(), 1, 128);
 
 	if (n > 1) {