perf util: Use cached rbtree for rblists

At the cost of an extra pointer, we can avoid the O(logN) cost of
finding the first element in the tree (smallest node), which is
something required for any of the strlist or intlist traversals
(XXX_for_each_entry()). There are a number of users in perf of these
(particularly strlists), including probes, and buildid.

Signed-off-by: Davidlohr Bueso <[e-mail]>
Tested-by: Arnaldo Carvalho de Melo <[e-mail]>
Cc: Jiri Olsa <[e-mail]>
Cc: Namhyung Kim <[e-mail]>
Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Arnaldo Carvalho de Melo <[e-mail]>

tools/perf/util/intlist.h

···
 /* For intlist iteration */
 static inline struct int_node *intlist__first(struct intlist *ilist)
 {
-	struct rb_node *rn = rb_first(&ilist->rblist.entries);
+	struct rb_node *rn = rb_first_cached(&ilist->rblist.entries);
 	return rn ? rb_entry(rn, struct int_node, rb_node) : NULL;
 }
 static inline struct int_node *intlist__next(struct int_node *in)

tools/perf/util/metricgroup.c

···
 	else if (metrics && !raw)
 		printf("\nMetrics:\n\n");
 
-	for (node = rb_first(&groups.entries); node; node = next) {
+	for (node = rb_first_cached(&groups.entries); node; node = next) {
 		struct mep *me = container_of(node, struct mep, nd);
 
 		if (metricgroups)

tools/perf/util/rb_resort.h

···
 
 /* For 'struct intlist' */
 #define DECLARE_RESORT_RB_INTLIST(__name, __ilist)				\
-	DECLARE_RESORT_RB(__name)(&__ilist->rblist.entries,			\
+	DECLARE_RESORT_RB(__name)(&__ilist->rblist.entries.rb_root,		\
 				  __ilist->rblist.nr_entries)
 
 /* For 'struct machine->threads' */

tools/perf/util/rblist.c

···
 
 int rblist__add_node(struct rblist *rblist, const void *new_entry)
 {
-	struct rb_node **p = &rblist->entries.rb_node;
+	struct rb_node **p = &rblist->entries.rb_root.rb_node;
 	struct rb_node *parent = NULL, *new_node;
+	bool leftmost = true;
 
 	while (*p != NULL) {
 		int rc;
···
 		rc = rblist->node_cmp(parent, new_entry);
 		if (rc > 0)
 			p = &(*p)->rb_left;
-		else if (rc < 0)
+		else if (rc < 0) {
 			p = &(*p)->rb_right;
+			leftmost = false;
+		}
 		else
 			return -EEXIST;
 	}
···
 		return -ENOMEM;
 
 	rb_link_node(new_node, parent, p);
-	rb_insert_color(new_node, &rblist->entries);
+	rb_insert_color_cached(new_node, &rblist->entries, leftmost);
 	++rblist->nr_entries;
 
 	return 0;
···
 
 void rblist__remove_node(struct rblist *rblist, struct rb_node *rb_node)
 {
-	rb_erase(rb_node, &rblist->entries);
+	rb_erase_cached(rb_node, &rblist->entries);
 	--rblist->nr_entries;
 	rblist->node_delete(rblist, rb_node);
 }
···
 					 const void *entry,
 					 bool create)
 {
-	struct rb_node **p = &rblist->entries.rb_node;
+	struct rb_node **p = &rblist->entries.rb_root.rb_node;
 	struct rb_node *parent = NULL, *new_node = NULL;
+	bool leftmost = true;
 
 	while (*p != NULL) {
 		int rc;
···
 		rc = rblist->node_cmp(parent, entry);
 		if (rc > 0)
 			p = &(*p)->rb_left;
-		else if (rc < 0)
+		else if (rc < 0) {
 			p = &(*p)->rb_right;
+			leftmost = false;
+		}
 		else
 			return parent;
 	}
···
 		new_node = rblist->node_new(rblist, entry);
 		if (new_node) {
 			rb_link_node(new_node, parent, p);
-			rb_insert_color(new_node, &rblist->entries);
+			rb_insert_color_cached(new_node,
+					       &rblist->entries, leftmost);
 			++rblist->nr_entries;
 		}
 	}
···
 void rblist__init(struct rblist *rblist)
 {
 	if (rblist != NULL) {
-		rblist->entries	 = RB_ROOT;
+		rblist->entries	 = RB_ROOT_CACHED;
 		rblist->nr_entries = 0;
 	}
 
···
 
 void rblist__exit(struct rblist *rblist)
 {
-	struct rb_node *pos, *next = rb_first(&rblist->entries);
+	struct rb_node *pos, *next = rb_first_cached(&rblist->entries);
 
 	while (next) {
 		pos = next;
···
 {
 	struct rb_node *node;
 
-	for (node = rb_first(&rblist->entries); node; node = rb_next(node)) {
+	for (node = rb_first_cached(&rblist->entries); node;
+	     node = rb_next(node)) {
 		if (!idx--)
 			return node;
 	}

tools/perf/util/rblist.h

···
  */
 
 struct rblist {
-	struct rb_root entries;
+	struct rb_root_cached entries;
 	unsigned int   nr_entries;
 
 	int (*node_cmp)(struct rb_node *rbn, const void *entry);

tools/perf/util/stat-shadow.c

···
 	struct rb_node *pos, *next;
 
 	rblist = &st->value_list;
-	next = rb_first(&rblist->entries);
+	next = rb_first_cached(&rblist->entries);
 	while (next) {
 		pos = next;
 		next = rb_next(pos);

tools/perf/util/strlist.h

···
 /* For strlist iteration */
 static inline struct str_node *strlist__first(struct strlist *slist)
 {
-	struct rb_node *rn = rb_first(&slist->rblist.entries);
+	struct rb_node *rn = rb_first_cached(&slist->rblist.entries);
 	return rn ? rb_entry(rn, struct str_node, rb_node) : NULL;
 }
 static inline struct str_node *strlist__next(struct str_node *sn)